Aexcel. Specialist Designation in Aetna Performance Network. Methodology Guide

Size: px
Start display at page:

Download "Aexcel. Specialist Designation in Aetna Performance Network. Methodology Guide"

Transcription

1 Aexcel Specialist Designation in Aetna Performance Network Methodology Guide

2 Aexcel 2009 TABLE OF CONTENTS Background on Aexcel Performance Networks... 3 Clinical Performance Evaluation Process... 8 General Statements on the Clinical Performance Measures... 9 Individual Clinical Measure Descriptions Adverse Event Rate (Measure ID 30100) Day Hospital Readmission Rate (Measure ID 30200)...16 Breast Cancer Screening (Measure ID 30600) Cervical Cancer Screening (Measure ID 30700) Testing for HIV during Pregnancy (Measure ID 30800) Lipid-Lowering Drug for Prevention of Ischemic Heart Disease (IHD) (Measure ID 30300) Beta-Blocker Treatment after AMI (Measure ID 30400) ACE/ARB Use in Members with Chronic Heart Failure (Measure ID 30500) Efficiency Evaluation Process General Statements on the Physician Efficiency Measure Episode of Care (EOC) Methodology Symmetry s as a Measure of Physician Group Efficiency Aetna s Provider Attribution for Symmetry Episodes of Care Provider Attribution for Aexcel Specialties Provider Minimum Episode Volume Symmetry Clean Period Claims Lag Episode Risk Groups Aetna Case-Mix Adjustment Aetna Outlier Logic Administrative Specification Provider Grouping Provider Scoring Statistical Significance of Efficiency Index Aexcel Designation Model Process Designation Process Appendix A: Aexcel Specialties and Sub Categories...39 Appendix B: Specialty Groupings Appendix C 2009 Aexcel Markets Copyright Aetna Inc. 2

3 Overview Background on Aexcel Performance Networks As one of the oldest and largest insurers in America, Aetna has an obligation and an opportunity to transform health care. We believe a better health care system is more transparent and consumer friendly and also recognizes physicians for their efficient and effective use of health care resources. Our Aexcel physician performance program is a component of our overall transparency efforts. Aexcel is a designation for specialists who: are part of the broader Aetna network of participating providers have met certain clinical performance and efficiency standards Aexcel originated from discussions with large employer groups who were challenged by rising health care costs. Patients, in turn, were becoming increasingly engaged as consumers of health care. As such, they wanted access to information about physicians to help them make informed health care decisions before seeking care from a physician. As Aexcel is implemented throughout the country, we make sure that affected physicians are aware of its introduction to their area beforehand. We also review the program with specialty societies and other groups within organized medicine. Our goal is to work collaboratively with physicians, employers and consumers to transform health care in a way that works for all constituents. We chose to address physician specialty care in developing this program for several reasons: 1. Specialty care is more episodic than primary care 2. Specialty care drives most of the advances in treatment, procedures, pharmaceuticals and diagnostic imaging, as well as the cost increases that accompany these advances 3. The following specialty categories chosen as part of Aexcel represent approximately 70 percent of specialty costs and control approximately 50 percent of our plan sponsors total medical costs Aexcel physician specialty categories Cardiology Neurology Otolaryngology/ENT Cardiothoracic surgery Neurosurgery Plastic surgery Gastroenterology Obstetrics and Gynecology Urology General surgery Orthopedics Vascular surgery Copyright Aetna Inc. 3

4 How We Evaluate Physicians Case Volume We begin by identifying specialists/groups currently participating in Aetna s network who have managed at least 20 episodes of care for Aetna members over the past 3 years. Only physicians with this baseline of episodes are considered for Aexcel designation. The next steps are to review a physician s: clinical performance efficiency Clinical Performance Clinical performance means how well physicians meet certain recognized measures of clinical care. We use clinical guidelines established by leading medical associations and considered part of standard clinical practice. These medical associations are well regarded by physicians and include the American College of Cardiology, American Heart Association, American College of Obstetricians and Gynecologists, Agency for Healthcare Research and Quality, Centers for Medicare & Medicaid Services, the Joint Commission, National Committee for Quality Assurance and Ambulatory Care Quality Alliance. Using claims information, we evaluate whether the physician met the clinical performance standards established by respected professional organizations. Some clinical performance measures pertain to all Aexcel specialties. Other measures are specialty specific. All measures are case-mix adjusted. All are associated with improved patient care. Metrics for all specialty categories include: A rate of unanticipated hospital readmissions within 30 days of discharge. A rate of unexpected adverse health events experienced by a specialist s hospitalized patients. Metrics for cardiologists only: A rate of use of cholesterol-lowering medication in patients with coronary artery disease. A rate of use of beta blocker in members identified with ischemic heart disease that has had a heart attack. A rate of ACE inhibitor or angiotensin receptor blocker in members with chronic heart failure. Metrics for obstetricians/gynecologists only: A rate of cervical cancer screening (Pap smear or similar test). A rate of breast cancer screening (mammography). A rate of HIV testing in pregnant women. To be evaluated for a clinical measure, physicians must have a minimum threshold of 10 cases per measure. The results of each measure are combined. Only those results that are statistically 95 percent confident or better are used in the Aexcel designation process. The clinical performance standards we use have already been adopted by many physicians as part of their standard clinical practice. For that reason, 96 percent of physicians evaluated for Aexcel pass these clinical performance measures. Copyright Aetna Inc. 4

5 Efficiency For physicians who meet the clinical performance standards, a measure of the efficiency of their care is developed and compared to their peers. We use the Symmetry Episode Treatment Groups methodology to measure a physician s efficiency. Efficiency is a combination of physician treatment patterns, health care resource utilization and cost. When evaluating the costs physicians incur for treating Aetna members, all costs are taken into account, not just office visits. We review inpatient, outpatient, diagnostic, laboratory and pharmacy costs for patients of these specialists. Statistical Validity For statistical validity, physicians must have a minimum of 20 episodes of care over 3 years. An index rating is created based on actual cost for the episode compared with the expected cost of the peer group. Episodes are then attributed to physicians. Surgical episodes are attributed to the surgeon with the highest allowed charges. If the episode is non-surgical, the physician with the highest number of visits receives the attribution of the case. Risk Adjustment Some physicians may care for more patients with chronic or complex conditions in a given time period than their peers; therefore, we evaluate physicians by comparing their services for patients with similar conditions. We apply risk-adjustment factors to account for differences in the use of health care resources among individuals. Use of health care resources can differ among patients because of age, gender, chronic disease risk and insurance product type. Comparison to Peers In addition, we compare all the resources used to treat a physician s patients to those of other physicians in the same specialty and geographic location. If a physician is a part of a group practice, we evaluate the entire group. In this case, performance measurement results of other physicians in the group practice will have an impact on each individual physician s evaluation. Episodes of Care This methodology is based on episodes of care, which is the current industry standard for measuring efficiency. Episodes of care methodology focuses on all of the costs (inpatient, outpatient, professional, office, lab, pharmacy and ancillary) required to care for a patient s underlying medical condition. Volume Threshold We chose 20 as a minimum threshold based on a comparison of results using random samples of various thresholds, including 20, 30 and 100. We found there was a reasonably similar result for groups with at least 20 episodes as there were at the higher thresholds. Furthermore, using 20 episodes as a minimum allows us to be more inclusive in our program. Application of Confidence Interval A physician whose efficiency score is greater than or equal to the mean efficiency score for his or her market and specialty is considered efficient. Physicians who are efficient, and statistically so, using a 90 percent confidence interval, are designated for Aexcel participation. In our experience, efficient physicians tend to recommend appropriate testing and treatment for their patients. What s more, these physicians may use some of the most advanced and costly procedures, pharmaceuticals, diagnostic imaging and technology but in a skilled and thoughtful manner. This results in high-quality outcomes for even the most complex cases, avoids complications and efficiently manages total medical costs. Copyright Aetna Inc. 5

6 Network Adequacy Once the selections are complete, we may need to supplement the network with additional physicians to ensure members have satisfactory access to enough specialists. However, only physicians who have met the clinical performance standards are eligible for consideration to supplement the network. Periodic Physician Re-Evaluation for Aexcel Designation We re-evaluate a physician s performance at least every two years. As a result, a physician s Aexcel designation status could change. Physicians who previously did not receive Aexcel designation may now meet the criteria. Similarly, physicians who are currently designated may lose their designation because they did not meet the volume, clinical performance or efficiency standards. This could be due to a physician s individual performance or because the overall performance of the physician s peers in his or her market, whom the physician is measured against, has improved. Regardless of whether a physician receives Aexcel designation, he or she remains a participating physician in Aetna s broader network. We realize that physicians, members and employers alike are impacted by changes to the composition of Aetna networks. We do our best to consider member and physician disruption during the Aexcel evaluation process. Data Variability As Aexcel continues to evolve we look for opportunities to further enhance our methodology and evaluation process based on new clinical evidence, feedback from members, providers and employers, as well as evolving industry trends. While we are committed to using the best available information, there are certain data limitations: The clinical quality and efficiency information is based on Aetna member data only. Combined claim data from multiple payors (e.g. insurance companies, self-insured and government plans) may provide a more complete picture of physician performance but is not yet available. We support industry-wide data collection initiatives and when this credible combined data becomes available, we will consider using it in our evaluations. The claim data used to evaluate physicians does not include all procedures, lab or pharmacy data - only those for which Aetna has claim data. Doctors may perform health care services for which they do not provide us with information. Also because of the way claims are submitted by doctors and/or processed by Aetna, health care service details may not always be available in the claim data we use. Therefore, we strongly encourage physicians to reach out to us with additional data they might have in medical charts that is not available to us through claims data. Aexcel as a Guide Aexcel information is intended to be only a guide for when a member chooses a specialist within the Aexcel specialist categories. There are many ways to evaluate doctor practices and members should consult with their existing doctor before making a decision. All ratings have a risk of error and, therefore, should not be the sole basis for selecting a doctor. We recommend members consider the following when selecting a doctor: Consult with their doctor regarding their health care decisions View specialist clinical quality and cost efficiency information as one factor in their health care decision Aexcel designation is not a guarantee as to the quality of the service a member receives or the outcome of any treatment by that specialist If a specialist is currently not designated for Aexcel: Copyright Aetna Inc. 6

7 o o o this does not mean the doctor does not provide quality services we might not have sufficient data to evaluate this doctor; and/or this doctor might be in the process of appealing their designation status Copyright Aetna Inc. 7

8 Clinical Performance Evaluation Process The Aexcel designation process includes four key criteria: Volume Clinical performance Efficiency and Network adequacy The following steps describe the clinical performance evaluation process. Step 1 All Aetna participating physicians in a geographic market 1 who practice in the selected specialty (for example, all cardiologists in the Aetna network in Atlanta) are reviewed. Physicians are rank-ordered according to an overall index score. Index measures are based on established evidence-based measures of clinical performance. Each measure is case-mix adjusted and a physician or physician group must have at least 10 cases in any given measure, for claims-based clinical performance measures evaluation. Clinical volume is based on a denominator of at least 10 in each measure used. The denominator can represent unique members or events, depending on the measure. In some measures, such as breast cancer screening, the denominator is members. In some measures, such as adverse event rate, the denominator is each event, and a member can have multiple events. Only scored measures are included in the index score; measures are weighted according to the number of eligible cases. Step 2 We identify physicians with the lowest index scores. Physicians whose measured outcomes fall below the 5 th percentile of the peer group are reviewed further (Steps 3-6) and may be excluded from consideration for Aexcel designation, unless other clinical criteria are met. Step 3 We apply a statistical significance formula (95 percent confidence limits) to the lowest group, removing any cases with insufficient statistical significance and reducing the group that may be excluded from Aexcel designation. Step 4 An Aetna medical director reviews measure detail reports of physicians remaining in the lowest group using available clinical data, including administrative claims data. Some cases have logical clinical explanations and are eliminated from the index score, allowing additional physicians to be considered for Aexcel designation. Step 5 Prior to any final decision about designation and publication of results, detailed clinical performance data for each measure is shared with the physicians remaining in the lowest group through Aexcel s formal reconsideration process. An Aetna medical director is available to discuss this data. Pursuant to the physician reconsideration process, every physician has the opportunity to provide additional information for reconsideration prior to the final clinical performance designation decisions. Step 6 Physicians who meet the clinical performance standards are evaluated on the efficiency of their care. On average, 96 percent of specialists evaluated pass these clinical performance measures. 1 In 2009, Aexcel is in 36 markets, including Arizona; Atlanta; Austin; Central Valley CA; Chicago, IL; Cincinnati; Cleveland; Colorado; Columbus; Connecticut; Dallas/Fort Worth; Delaware; Detroit; Houston; Indianapolis; Kansas City, KS and MO; Los Angeles; Louisville; Maine; Metropolitan DC (including Maryland, DC and Northern Virginia); Metropolitan New York; North Florida; Northern CA; Northern New Jersey; Oklahoma City; Orlando; Pittsburgh; Richmond; San Antonio, TX; San Diego; Seattle/W.Washington; South FL (Dade and Broward Counties); Tampa; and Tulsa. Copyright Aetna Inc. 8

9 General Statements on the Clinical Performance Measures 1. Current Procedure Terminology All references are to Current Procedure Terminology 2007 American Medical Association. All rights reserved. 2. Provider Attribution Attribution logic is addressed for each specific measure. The Enterprise Provider Database (EPDB) is a database that houses provider and network data. EPDB houses two types of provider data: community and network participation data. Community data is information that is generally known about a Provider. Examples are name, Social Security number, tax identification number, service location(s), billing address(es), education. Provider network participation data (PNP) is information specific to a provider's contract or participation with Aetna. Examples are fees, fee codes, risk groups, effective and expiration network participation dates, network specialties, directory information. Only participating providers are considered for Aexcel. Twelve specialties (cardiology, cardiothoracic surgery, gastroenterology, general surgery, obstetrics and gynecology, orthopedics, otolaryngology, neurology, neurosurgery, plastic surgery, urology and vascular surgery) are included within the Aexcel network, and the appropriate specialty is addressed within the measurement specifications. For a complete list of sub-specialties refer to Appendix A. Visit logic: The Aetna Data Warehouse Procedure Group contains the historic and current evaluation and management (physician visit) codes as well as CPTII codes that give assessment information on a member gathered during a physician visit, and HCPC codes that indicate physician visits for specific care. These procedure codes represent contact with the member where a physical assessment was completed by a physician provider. We have not limited the scope of the evaluation and management codes included (assessments in the outpatient, ER or inpatient settings), but rather have limited the attribution of the visit based on the place of service on the claim or encounter. Each measure defines the specialty that is included for attribution. Within EPDB, each provider/provider group is assigned a provider type and a specialty. We have grouped the codes that represent the specific specialties to define provider attribution. 3. Lab Data Integrity We have found that a data integrity issue exists for certain areas of the country due to capitation arrangements with labs, where we are not receiving all the encounters for the labs that are done. For HMO-based membership, a logical method is applied at the provider level to determine laboratory utilization. If a provider qualifies for data integrity condition, which is low utilization, all attributed members for that provider are excluded at the individual ID level. (Traditional plan data is not subject to the same problem since we receive the claims for those tests.) 4. Claims Lag Four months of medical claim lag is required. We add four months on to the end of the measurement timeframe before we extract the data from our warehouse. This ensures the majority of medical claims have been submitted to be included in our data. Pharmacy claims are current due to electronic filing, so they are retrieved according to Copyright Aetna Inc. 9

10 the timeframe in the measure for the pharmacy claims. Pharmacy claims can be extracted as soon as the end of month data is loaded into the warehouse or later if the end date is used to limit the timeframe. 5. Case Mix Adjustment Case mix classifies data characteristics into groups that are homogenous to allow for a basis of comparison. A case-mix adjustment is applied to each clinical performance measure, as appropriate, such as the ob/gyn and cardiology clinical performance measures, for which there is strong evidence in the literature of the value of certain services and medications for all members meeting the criteria of each measure. For this reason, it was not necessary to adjust provider results to account for differences in the age, gender and severity level of their patient population. However, because the measurement for each provider includes all of his/her Aetna members regardless of the type of medical product (HMO-based or traditional-based), we did casemix adjust the results by the types of medical products of the Aetna members in the provider s panel. We were concerned that incomplete capture of capitated services (encounters) for HMO-based plan members could negatively impact providers who had a greater proportion of Aetna HMO-based plan members in their patient population compared to providers with a greater proportion of Aetna traditional-based members in their patient population. For consistency in programming, the medical product adjustment was applied for all of these measures, although we did not expect capitation to affect pharmacy data. We also adjusted rates of breast cancer screening by plan market (a geographic variable) because of reports of prolonged scheduling delays at mammography centers in certain markets within Aetna regions. Such delays could affect providers differently when compared to a regional or national average. The expected value for all measures, except adverse event, is calculated based on the results of indirect standardization using case-mix adjustment. The expected value for adverse event is calculated using regression analysis. 6. Medical Case Aetna Medical Case Logic summarizes clinical events by linking or associating all of the claims submitted for a member during the same treatment episode. All Aetna Data Warehouse claims and encounter details are run through medical case logic. All specialist and ancillary claims that are within the starting and ending dates of service for these cases are attached to the case. For inpatient cases, a room and board bill is required based on the revenue code or benefit code. A bill from a different facility would trigger a new case. An example would be a transfer or being moved from a medical bed to a skilled nursing or rehab bed. Once a case is defined, a series of clinical attributes are defined for the case. For example, each medical case is assigned a Diagnostic Related Group (DRG) to aid in the analysis. A DRG is assigned to each case using DRG Grouper software. The array of codes fed to the Grouper is arranged based on a clinical algorithm and service date basis, thus delivering the most clinically appropriate DRG for the case. Aetna has defined algorithms that identify a managing provider for each case. Copyright Aetna Inc. 10

11 7. Inpatient Performance Measurement System The Aetna Inpatient Performance Measurement System (IPMS) is a case-mix and severity-adjusted inpatient care performance measurement tool that Aetna uses in a number of ways, one of which is Aexcel. IPMS obtains its original data from medical, pharmacy and lab claims, as well as from member and provider data. Factors that are not under the control of the hospital have an impact on length of stay and the likelihood of having an adverse event. Examples of patient-specific factors that impact outcomes include age, gender, admission type (elective, emergency, transfer from another facility, etc.), and comorbidities (clinical conditions present at the time of hospital admission). IPMS applies a regression-based methodology to the inpatient data to account and adjust for these patient-specific factors. With this approach, adjusted expected rates can be calculated. Expected rates can then be compared with observed rates. This comparison of observed to expected provides a valid measure of performance specific to the population of interest. And, since the calculated data has been adjusted for clinical and demographic factors, an equitable comparison can be made across hospitals, providers or geography. 8. Health Profile Database The Aetna Health Profile Database (HPD) is a foundation database used to identify Aetna members with any of 81 chronic diseases or medical conditions. The identification algorithms are comprised of medical, pharmacy and clinical laboratory data from physician claims and encounters, specialist claims, pharmacy, facilities, laboratories and others. The HPD is refreshed monthly. The HPD consists of two major components: Disease Identification and Prevalence The chronic diseases and medical conditions included in HPD are selected based on the following considerations: 1.) the disease is chronic in nature, 2.) the disease represents a significant burden of illness, and 3.) the disease generates significant medical costs. In addition, diseases were chosen where improved processes of care might lead to better outcomes. Clinical selection criteria have been established for each disease in HPD. The Aetna Clinical Groups is used in the generation of these criteria. These groups are a hierarchical grouping of diagnosis and procedure codes used for classifying all International Classification of Diseases 9th revision (ICD-9) and Current Procedure Terminology 4th revision (CPT-4) codes. Complex inclusion and exclusion logic has been developed to reduce the occurrence of a false positive rate on specific indicators. Members are selected for inclusion in one or more HPD disease categories using one hit criteria or one occurrence of any of the specific diagnosis criteria, including ICD-9, CPT, pharmacy and laboratory codes. This selection process is designed to loosely identify members for the calculation of prevalence of chronic diseases and clinical conditions in a population and to identify a pool of members for more rigorous identification and ad hoc reporting. Disease Indicator Flag A more stringent identification criterion for members with chronic diseases and clinical conditions is beneficial for targeted interventions, case management and member education through disease management programs. The use of a multiple selection criteria improves the sensitivity of the identification process. DM=Y Members flagged with a disease identification indicator of Y have fully met the disease-specific identification criteria defined as hits in two of five databases OR a total of three hits in any of the databases, except where exclusion logic applies. One date of service must fall within the prior 18 months. The 18-month timeframe is calculated from the file effective date. Copyright Aetna Inc. 11

12 DM=N Members flagged with a disease identification indicator of N have not fully met the identification criteria. This category may potentially be useful for early identification of chronic diseases. DM=H Members flagged with a disease identification indicator of H were at one time identified but have either termed or no longer meet the criteria for identification, for example, no date of service in the last 18 months. Use of HPD in Clinical Measures: For members to be included in a specific measure, the DM=Y or H is used to ensure that the member is being treated for the specific disease or condition. For measures that require member exclusion based on knowledge of another disease or condition, the existence of that disease code in the HPD, which is equivalent to a single claim for that condition, is considered evidence of the excluded value. 9. Clinical Group Definitions UPG, or Aetna Procedure Groups, represent the grouping of similar CPT 4 and ICD 9 procedure codes that refer to specific procedures performed for similar pathophysiologic processes for the same organ system. Currently, there are over 26,000 procedure codes, which map into 191 internally defined procedure groups or UPGs. Each procedure group is categorized as to whether it is major procedure (for example, heart surgery), a minor procedure (for example, colonoscopy), a diagnosis or other. UDG, or Aetna Diagnosis Groups, represent a grouping of similar ICD 9 diagnosis codes that refer to the same pathophysiologic process affecting the same organ system. Currently, there are over 16,000 diagnosis codes, which map into 198 diagnosis groups, or UDGs. CSG, Clinical Service Group, is an Aetna grouping that reflects the most significant clinical reason for a medical case; it can be a procedure or a diagnosis. Which one is assigned to the CSG is dependent on the place of service, procedures and diagnoses submitted on the claims that linked to that case. IPMS Diagnostic Related Group (IDRG) is an aggregate of HCFA DRGs that represent the outputs of reasonably distinct clinical processes. The IDRG logic is used in IPMS adverse-event and length-of-stay modeling to group the DRGs into similar subsets based on the member s critical path (clinical reason for hospitalization) and then into the IDRG specialty for modeling. Copyright Aetna Inc. 12

13 Individual Clinical Measure Descriptions Adverse Event Rate (Measure ID 30100) Measure Endorsed by: AHRQ/RAND Description A case-mix adjusted rate of unexpected adverse health events experienced by a specialist s hospitalized patients. The adverse event data for the Aexcel measure is extracted from the Inpatient Performance Measurement System (IPMS). Understanding Adverse Events An adverse event is a negative, unanticipated consequence of care. Examples include a wound infection within an elective surgery case, hospital-acquired pneumonia or deep vein thrombosis after hip replacement surgery. All cases are included in the modeling for adverse events; however, not all of these are included in the Aexcel specialist measurement. Only those deemed clinically appropriate are used in Aexcel. A subset of adverse events related to anesthesia and drug-induced complications have been excluded from this measure because these are complications that are not likely controlled by care directed by the managing provider of the medical case. IPMS logic assigns an adverse event to a case if there is a secondary diagnosis that has all of the following characteristics: The diagnosis is unlikely to have been present on admission. The diagnosis is unlikely to be a coexisting condition. The diagnosis is unlikely to reflect the normal progression of the principal diagnosis. The diagnosis is logically consistent with the occurrence of an adverse event. Modeling Adverse Events Cases are grouped by DRG to align with provider practice specialties, such as cardiology, cardiothoracic surgery, gastroenterology and general surgery. Each group of DRG cases is modeled based on the specialty definition. Each case is then interrogated for secondary diagnosis codes that Aetna has targeted as representing a possible adverse event. The case is checked to see if it is an elective admission or an emergency admission. Due to the known instability of members admitted for urgent care, some adverse event ICD-9 codes are only considered for elective admissions. Some complications are anticipated, such as the member who has cardiac surgery but goes on to have a postoperative heart attack. This ICD-9 diagnosis code would not be counted as an adverse event. The exclusion logic, organized by specialty model, excludes the count for adverse event diagnosis codes present in an admission, based on the IDRG of the case, to avoid the inappropriate counting of complications that are unfortunate but anticipated. Copyright Aetna Inc. 13

14 Eligible Population Clinical Subset The CSG is an Aetna-derived grouping that reflects the most significant clinical reason for a medical case. The CSG subset is used to identify the most appropriate inpatient cases to be evaluated based on the provider s specialty using the diagnosis and procedure codes that the specialist is expected to manage. Continuous Enrollment Measurement Period Benefit Provider Specialty Eligibility based on an inpatient stay 24 months Medical There are currently 14 IPMS specialties that are modeled along the adverse event metric, and they are: cardiology, cardiothoracic surgery, gastroenterology, general medicine, general surgery, gynecology, hematology/oncology, neurology, neurosurgery, obstetrics, orthopedics, otolaryngology, pulmonary medicine and urology. General medicine is split into adults and pediatrics. Administrative Specification Denominator The acute inpatient cases identified through the IPMS process with an inpatient stay. When we use the adverse event measure in Aexcel, we only use the events that have a managing specialist in one of the following specialty categories: cardiology, cardiothoracic surgery, gastroenterology, general surgery, obstetrics and gynecology, orthopedics, otolaryngology, neurology, neurosurgery, plastic surgery, urology and vascular surgery. Numerator Expected The numerator is an adverse event as described above. Separate data files are created for each specialty, clinical logic is applied and logistic regression models are run to obtain parameter estimates for the adverse event measure. An adjusted average rate is generated and is expressed as the expected adverse event rate on every medical case. The expected value is calculated based on the results of case-mix regression modeling that take into account many clinical and demographic variables that include, but are not limited to age, gender, region, product, admission type, discharge status, co-existing condition and the DRG group. A p value is generated and applied if statistically significant. Copyright Aetna Inc. 14

15 Exclusion The exclusion logic excludes the count for adverse event diagnosis codes present in an admission, based on the IDRG of the case, to avoid the inappropriate counting of complications that are unfortunate but anticipated. Provider Attribution Assigning the Managing Provider In assigning a managing provider of a medical case, or the physician who managed the clinical and diagnostic events of the case, all medical, surgical or Ob/Gyn physicians from the claims that feed into the case are considered. The managing provider audit code indicates how the managing provider for that case was determined. Possible values, in priority order of selection, are: 1 = The provider is a medical, surgical or Ob/Gyn physician that performed a major procedure (if applicable). 2 = The provider is a medical, surgical or Ob/Gyn physician that performed a minor procedure (if applicable). 3 = The provider is a medical, surgical or Ob/Gyn physician that has the highest number of visits. If the managing provider s specialty category falls into one of the following, then that provider s record is used to produce the clinical adverse event measure used in Aexcel: cardiology, cardiothoracic surgery, gastroenterology, general surgery, obstetrics and gynecology, orthopedics, otolaryngology, neurology, neurosurgery, plastic surgery, urology and vascular surgery. Observations A minimum of 10 eligible inpatient events, per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case Mix Adjustment An adjusted average rate is generated and is expressed as the expected adverse event rate. Every case generates this expected or predicted probability. The expected value is calculated based on the results of casemix regression modeling that takes into account many clinical and demographic variables that include, but are not limited to, age, gender, region, product, admission type, discharge status, co-existing condition and the DRG group. A p value is generated and applied if statistically significant. Copyright Aetna Inc. 15

16 30-Day Hospital Readmission Rate (Measure ID 30200) Measure Endorsed by: AHRQ/RAND Description A case-mix adjusted rate of unanticipated hospital readmissions within 30 days of discharge. This measures the readmission for a specialist or specialty group. Readmission is defined as the proportion of hospitalizations managed by the specialist or specialty group that are followed by a subsequent hospitalization related to the same condition or a recognized complication of the condition within 2 to 30 days of the discharge date of the first hospitalization. This measure excludes readmissions that would have been expected based on the clinical nature of the case. The actual readmission rate is compared with the adjusted average readmission rate for the specialist or specialty group. The average level of performance is adjusted for the age and gender of the member, as well as for the reason care was provided (defined by the diagnosis or procedure that best describes the reason for the admission). Eligible Population Product lines Ages Clinical Subset Continuous Enrollment Benefit Visit Provider Specialty All products (commercial, Medicare and Medicaid). All The CSG is an Aetna derived grouping that reflects the most significant clinical reason for a medical case. The CSG subset is used to identify the most appropriate inpatient cases to be evaluated based on the provider s specialty using the diagnosis and procedure codes that the specialist is expected to manage During the month of the inpatient admission through the month after discharge Medical Attribution is done according to managing provider logic for Aetna medical case grouper. The definition is according to greatest number of visits, unless surgical, in which case it is the surgeon. Any of the 12 Aexcel specialties that have an acute inpatient admit that fit the categories for this measure. Copyright Aetna Inc. 16

17 Administrative Specification Denominator Numerator All acute inpatient cases occurring within the first 23 months of the 2-year measurement period using discharge date, managed by a specialty provider with an Aexcel specialty to CSG mapping defined above. If an inpatient hospitalization occurred within 2 to 30 days of an inpatient discharge (denominator), it is considered a readmission. Exclusion Calculate the 99 th percentile based on Specialty and length of stay of the cases selected for the denominators. If the case is in the 99 th percentile the case will be excluded from the expected calculation, and a record will be produced for this case without an expected value. Provider Attribution Managing provider, according to medical case logic, on the initial admission. Observations A minimum of 10 eligible inpatient events per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case Mix Adjustment This measure is case-mix adjusted by age, product and CSG logic with specialty included. Copyright Aetna Inc. 17

18 Breast Cancer Screening (Measure ID 30600) Measure Endorsed by: NQF, AQA, AMA,/CMS/HEDIS/NCQA Description The percentage of women years of age who had a visit to an Ob/Gyn during the most recent 12-month measurement time period and who had a mammogram to screen for breast cancer. Eligible Population Product lines Ages Continuous enrollment Measurement Period Benefit Visit Provider Specialty All products (commercial, Medicare and Medicaid). Women years as of the end of the measurement period. Twelve months prior to the measurement end date. Twenty-four-month measurement time period to identify the numerator. Medical. Ob/Gyn provider during the measurement period, as defined by specialty category code (OG). One visit to an Ob/Gyn provider during the measurement time period Administrative Specification Denominator Numerator The eligible population. One mammogram during the 24-month measurement time period. A woman had a mammogram if a submitted claim/encounter contains any one of the codes in the following table. Table Measure ID A: Codes to Identify Breast Cancer Screening CPT HCPCS ICD-9-CM Diagnosis ICD-9-CM Procedure UB Revenue G0202, G0204 or G0206 V76.11, V , or 403 Copyright Aetna Inc. 18

19 Exclusion If the member had unilateral mastectomies on two separate dates of service or a bilateral mastectomy. All date frames available in the Aetna Data Warehouse are reviewed (three full years of history plus the current year). Refer to the following table for codes to identify exclusions. Table Measure ID B: Codes to Identify Exclusions Description CPT ICD-9-CM Procedure Bilateral mastectomy 19216, 19245, , 19187, 19205, or ,85.44, 85.46, 85.48, or Unilateral mastectomy (members must have 2 separate occurrences at least 14 days apart) 19180, 19200, 19220, 19240, 19250, 19254, 19215, 19182, or , 85.31, 85.33, 85.34,85.4, 85.41, 85.43,85.45 or Provider Attribution Each Ob/Gyn provider the member has visited, at least once, in the current measurement time period is included in the measure. The member may or may not have had the numerator service with a qualifying denominator provider. The measure evaluates each qualifying denominator provider whom the member visited. Only one provider per group, per member is attributed. The visit date selected is the most recent. Duplications of the same provider and member combination are eliminated. Observations A minimum of 10 eligible members per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case-Mix Adjustment Case-mix adjustment is applied to each clinical performance measure, as appropriate, and classifies data characteristics into groups that are homogenous to allow for a basis of comparison. Copyright Aetna Inc. 19

20 Cervical Cancer Screening (Measure ID 30700) Measure Endorsed by: NQF, AQA, HEDIS, NCQA Description The percentage of women years of age who had a visit to an Ob/Gyn during the most recent 12-month measurement time period and received a Pap test to screen for cervical cancer. Eligible Population Product lines Ages Continuous enrollment Measurement Period Benefit Visit Provider Specialty All products (commercial, Medicare and Medicaid). Women years as of the end date of the measurement time period. Twelve months prior to the measurement end date. Thirty-six-month measurement time period to identify the numerator. Medical. One visit to an Ob/Gyn provider during the measurement time period Ob/Gyn provider during the measurement period, as defined by specialty category code (OG). Administrative Specification Denominator Numerator The eligible population. One or more Pap tests during the measurement year or the two years prior to the measurement year. A woman had a Pap test if a submitted claim/encounter contains any one of the codes in the following table. Table Measure ID A: Codes to Identify Cervical Cancer Screening CPT HCPCS ICD-9-CM Diagnosis ICD-9-CM Procedure 88141, 88142, 88143, 88144, 88145, 88147, 88148, 88150, 88152, 88153, 88154, 88155, 88156, 88158, 88164, 88165, 88166, 88167, or G0214, G0123, G0141, G0143, G0144, G0145, G0147 or G0148 v76.2, v72.3 or v Copyright Aetna Inc. 20

21 Exclusion Exclude women who had a hysterectomy at any time, based on claims or encounters where the procedure code indicates a hysterectomy was performed. Refer to the following table to identify exclusions. Table Measure ID B: Codes to Identify Exclusions Description Health Profile Database Hysterectomy Excluded is procedure group number of 39 (hysterectomy) Provider Attribution Each Ob/Gyn provider the member has visited at least once in the current measurement time period is included in the measure. The member may or may not have had the numerator service with a qualifying denominator provider. Each qualifying denominator provider the member has visited will be evaluated. Only one provider per group, per member is attributed. The visit date selected is the most recent. Duplications of the same provider and member combination are eliminated. Observations A minimum of 10 eligible members per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case-Mix Adjustment Case-mix adjustment is applied to each clinical performance measure, as appropriate, and classifies data characteristics into groups that are homogenous to allow for a basis of comparison. Lab Data Integrity Lab data integrity applies. Copyright Aetna Inc. 21

22 Testing for HIV during Pregnancy (Measure ID 30800) Measure Endorsed by: NQF, AQA, ACOG Description The percentage of women greater than 10 years of age, who delivered a baby, had an HIV test performed, and a claim was received from an obstetrician who billed an ante partum visit or total obstetrical care during the measurement time period. Eligible Population Product lines Ages All products (commercial, Medicare and Medicaid). Women over 10 years of age as of the end date of the measurement period. Delivery Continuous enrollment Measurement Period Benefit Women who had a delivery during the measurement period, based on medical case delivery logic that defines a vaginal or Cesarean delivery. This value is determined based on procedure codes or diagnosis codes within the medical case that indicate a confinement for a delivery. The month of the delivery plus the nine months prior. The month of the delivery plus the nine months prior. Medical. Administrative Specification Denominator Numerator The eligible population. HIV testing within the 10-month measurement period. A woman had an HIV test if a submitted claim/encounter contains any one of the codes in the following table Table Measure ID A: Codes to Identify HIV Testing CPT , 86689, 87390, 87391, ,87901 Copyright Aetna Inc. 22

23 Exclusion Exclude from the denominator women who did not have a qualifying test performed and identified with AIDs Table Measure ID B: Codes to Identify Exclusions Description Health Profile Database AIDS Excluded are members with a disease code of AID. Provider Attribution Each Ob/Gyn provider who billed the total obstetric care or ante-partum care in the current measurement time period is included in the measure. The member may or may not have had the numerator service with a qualifying denominator provider. Each qualifying denominator provider who billed the total obstetric care or ante-partum care will be evaluated. Table Measure ID C: Codes to Identify Total OB Care CPT 59400, , 59510, or Only one provider per group, per member is attributed. If there are multiple deliveries for one member within a time period, each delivery is evaluated. Observations A minimum of 10 eligible members per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case-Mix Adjustment Case-mix adjustment is applied to each clinical performance measure, as appropriate, and classifies data characteristics into groups that are homogenous to allow for a basis of comparison. Lab Data Integrity Lab data integrity applies. Copyright Aetna Inc. 23

24 Lipid-Lowering Drug for Prevention of Ischemic Heart Disease (IHD) (Measure ID 30300) Measure Endorsed by: NQF, AQA, PCPI, ACC Description Measures the use of cholesterol-lowering medication in patients with ischemic heart disease (IHD). This measure includes only members with an Aetna pharmacy plan and evidence of use. Eligible Population Product lines Ages Diagnosis All products (commercial, Medicare and Medicaid). Members 22 years and older at the end of the measurement time period. Evidence of IHD on or before the visit to the cardiologist. Table Measure ID A contains codes. Table Measure ID A: Codes or Descriptive to Identify IHD and AMI Description Health Profile Database Ischemic Heart Disease Members with a disease code of IHD (DM = Y or H) Continuous Enrollment Based on the data source of the membership file, if the visit to the cardiologist is within the first 15 days of the month, then continuous enrollment starts the month prior to the cardiology visit and subsequent 4 months. If the visit to the cardiologist is after the 16 th of the month, then continuous enrollment starts the month of the cardiology visit and subsequent 4 months. Measurement Period Benefit Visit Provider Specialty Twenty-four months Medical and pharmacy. The pharmacy benefit must be demonstrated through an Aetna benefits indication during the month of the cardiology visit AND the 4 months following the cardiology visit. In addition, a pharmacy claim must have been received at any time during the 24- month reporting period or 4-month post-reporting period. A minimum of one visit to a cardiologist during the measurement time period Cardiologist provider during the measurement period, as defined by specialty category code (C) Copyright Aetna Inc. 24

25 Administrative Specification Denominator Numerator The eligible population. Member in the denominator with a pharmacy claim for a drug with one of the following descriptions. Table Measure ID B: Lipid Lowering Medications Description Antihyperlipidemics Niacin Exclusion No exclusions. Provider Attribution Each cardiologist the member has visited, at least once, in the current measurement time period is included in the measure. The member may or may not have had the numerator service with a qualifying denominator provider. Each qualifying denominator provider the member has visited will be evaluated. Only one provider per group, per member is attributed. Observations A minimum of 10 eligible members per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case-Mix Adjustment Case-mix adjustment is applied to each clinical performance measure, as appropriate, and classifies data characteristics into groups that are homogenous to allow for a basis of comparison. Copyright Aetna Inc. 25

26 Beta-Blocker Treatment after AMI (Measure ID 30400) Measure Endorsed by: AQA, HEDIS/NCQA Description Measures the use of beta blocker in members identified with ischemic heart disease who have had a heart attack. This measure includes only members with an Aetna pharmacy plan and evidence of use. Eligible Population Product lines Ages Diagnosis All products (commercial, Medicare and Medicaid). Members 22 years and older at the end of the measurement time period. Evidence of AMI on or before the visit to the cardiologist. Table Measure ID A contains codes. Table Measure ID A: Codes or Descriptive to Identify IHD and AMI Description Health Profile Database Ischemic Heart Disease Members with a disease code of IHD (DM = Y or H) Acute Myocardial Infarction Members with AMI indicator of yes Continuous Enrollment Benefit Based on the data source of the membership file, if the visit to the cardiologist is within the first 15 days of the month, then continuous enrollment starts the month prior to the cardiology visit and subsequent 4 months. If the visit to the cardiologist is after the 16 th of the month, then continuous enrollment starts the month of the cardiology visit and subsequent 4 months. Medical and pharmacy. The pharmacy benefit must be demonstrated through an Aetna benefits indication during the month of the cardiology visit AND the 4 months following the cardiology visit. In addition, a pharmacy claim must have been received at any time during the 24-month reporting period or 4- month post-reporting period. Copyright Aetna Inc. 26

27 Administrative Specification Denominator Numerator The eligible population. Member who has a pharmacy claim for a drug with one of the following descriptions. Table Measure ID B: Medications Description Beta blockers non selective Beta blockers cardio selective Alpha-beta blockers (Coreg/Carvedilol only) Exclusion Exclude from the denominator members with any of the disease or ICD-9 codes in Table Measure C found within the timeframes available within Aetna s data warehouse. Members with a diagnosis of asthma, COPD or chronic renal failure are excluded. Table Measure ID C: Exclusion Codes ICD- Description ICD 9 Codes Atrioventricular Block, Complete 426 Atrioventricular Block, Other And Unspecified Mobitz (type) Ii Atrioventricular Block Other Second Degree Atrioventricular Block Other Heart Block Sinoatrial Node Dysfunction Table Measure ID D: Exclusion by Disease Description Health Profile Database Asthma Members with a disease code of AST COPD Members with a disease code of COP Chronic Renal Failure Members with a disease code of CRF Provider Attribution Each cardiologist the member has visited, at least once, in the current measurement time period is included in the measure. The member may or may not have had the numerator service with a qualifying denominator provider. Each qualifying denominator provider the member has visited will be evaluated. Only one provider per group per member is attributed. Copyright Aetna Inc. 27

28 Observations A minimum of 10 eligible members per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Case Mix Adjustment Case mix adjustment is applied to each clinical performance measure, as appropriate, and classifies data characteristics into groups that are homogenous to allow for a basis of comparison. Copyright Aetna Inc. 28

29 ACE/ARB Use in Members with Chronic Heart Failure (Measure ID 30500) Measure Endorsed by: NQF, AQA, PCPI, CMS/JCAHO Description Measures the rate of ACE inhibitor or angiotensin receptor blocker (ARB) in members with chronic heart failure. This measure includes only members with an Aetna pharmacy plan and evidence of use. Eligible Population Product lines Ages Diagnosis All products (commercial, Medicare and Medicaid). Members 20 years and older as of the end of the measurement time period. Evidence of CHF on or before the visit to the cardiologist. Table Measure ID A contains codes. Table Measure ID A: Codes to Identify CHF Description Health Profile Database Congestive Heart Failure Members with a disease code of CHF (DM = Y or H) Continuous Enrollment Benefit Visit Based on the data source of the membership file, if the visit to the cardiologist is within the first 15 days of the month, then continuous enrollment starts the month prior to the cardiology visit and subsequent 4 months. If the visit to the cardiologist is after the 16 th of the month, then continuous enrollment starts the month of the cardiology visit and subsequent 4 months. Medical and pharmacy. The pharmacy benefit must be demonstrated through an Aetna benefits indication during the month of the cardiology visit AND the 4 months following the cardiology visit. In addition, a pharmacy claim must have been received at any time during the 24-month reporting period or 4-month post-reporting period. A minimum of one visit to a cardiologist during the measurement time period Provider Specialty Cardiologist during the measurement period, as defined by provider specialty category code (C) Copyright Aetna Inc. 29

30 Administrative Specification Denominator Numerator The eligible population. Member who has a pharmacy claim for a drug with one of the following descriptions. Table Measure ID B: Medications Description of Medications Carvedilol Bisoprolol fumarate Metoprolol Ace Inhibitors Angiotensin II receptor antagonists ACE Inhibitors and Calcium Channel Blockers ACE Inhibitors and thiazides Angiotensin II receptor antagonists and thiazides OR Hydralazine and Nitrates OR Vasodilators/Thiazide and Nitrates Exclusion Exclude from the denominator members with any of the individual ICD-9 codes in Table Measure C or evidence of an ICD-9 diagnosis code within any of the listed diagnosis groups. Table Measure ID C: Codes to Identify Exclusions Description ICD-9 Codes Hypertrophic Obstructive Cardiomyopathy Hyperkinetic Heart Disease Table Measure ID D: Codes to Identify Exclusions Description Diagnosis Group Number Congenital Heart disease 7 Acute Renal Failure 66 Chronic Renal Failure 67 Provider Attribution Each cardiologist the member has visited, at least once, in the current measurement time period is included in the measure. The member may or may not have had the numerator service with a qualifying denominator provider. Each qualifying denominator provider the member has visited will be evaluated. Only one provider per group, per member is attributed. Observations A minimum of10 eligible members, per provider/provider group, as defined by the denominator, is needed for a provider to be scored on this measure. Copyright Aetna Inc. 30

31 Case Mix Adjustment Case mix adjustment is applied to each clinical performance measure, as appropriate, and classifies data characteristics into groups that are homogenous to allow for a basis of comparison. Copyright Aetna Inc. 31

32 Efficiency Evaluation Process General Statements on the Physician Efficiency Measure Episode of Care (EOC) Methodology Episode of care is a methodology to assist in understanding medical cost and utilization drivers. An "episode of care" for a member represents treatment and service utilization across time for an identified health condition. The doctor, hospital, pharmacy and ancillary testing, as well as other costs and utilization relating to an episode of illness, are rolled up into a single entity based on a specific condition. An episode of care spans from the onset of symptoms until treatment is complete. Aetna uses Symmetry Health Data Systems software and its Episode Treatment Groups (s ) version 6.5 illness classification system to build episodes of care data in the Aetna Data Warehouse. The technology is distributed in the form of "grouper" software. The software accepts health care claims (service line detail) and returns the value, along with other patient information. The grouper software rolls up all doctor, hospital, pharmacy and ancillary testing claims data together to create episodes of illness or surgery that are clinically defined and identified by codes for analysis purposes. Symmetry s as a Measure of Physician Group Efficiency Symmetry s s offer the ability to identify, quantify and compare the total medical costs of a clinically based episode of care spanning hospitalizations, ambulatory visits and all ancillary services, including the use of pharmaceuticals. Medical claims and pharmacy claims are fed into the grouper software, and medical episodes are created from these claims. Symmetry creates s that categorize the episodes into different medical groupings. s are assigned to a physician who is involved in a patient s care as demonstrated by that provider s claims detail within that episode. The physician who is assigned the episode is referred to as the responsible provider. Aetna has developed the responsible provider logic that is used to assign an episode of care to a physician. There are many variables that can affect the use of health care resources for a condition. There can be variation in resource use to treat a condition that is a direct result of the population differences or level of illness. To adjust for the variation in resource use, a case-mix-adjusted expected allowed amount is created for each that is assigned to a physician group. That case-mix-adjusted expected allowed amount is then compared to the actual allowed amount for that episode of care. The efficiency measure created for Aexcel provides the ability to evaluate medical cost and utilization patterns among physicians treating similar, severity-adjusted illnesses. Providers are ranked in terms of resource efficiency after adjusting for differences in patient demographics and case mix, and then are compared to their peers. The peer group is defined as physicians of the same specialty in the same Aexcel market. There can be variation in efficiency indexes over time period analysis. Variation has been noted in physician groups whose episode counts over the given time period are low. To evaluate physicians and determine potential for variation, the statistical significance of a provider group s efficiency score is evaluated using a 90 percent confidence interval. Copyright Aetna Inc. 32

33 Aetna s Provider Attribution for Symmetry Episodes of Care Aetna assigns one responsible provider for each episode of care. For non-surgical s, an episode is assigned to the physician with the most visits with the patient during that episode of care. For surgical s, an episode is assigned to the surgeon with the most allowed amounts for that episode. Only one physician is assigned to an episode with Symmetry. Provider Attribution for Aexcel Specialties Aexcel uses the responsible provider of the episode for attribution but limits the s used for evaluation of efficiency in the Aexcel network selection process. Each specialty has a specialty specific list of s that are used in the evaluation process, and they are limited to the s that are most frequently assigned to that particular specialty (see Appendix B). Upon request, Aetna will discuss with physicians more detailed information regarding the s that were managed by the physicians. Provider Minimum Episode Volume We identify those physicians who have managed at least 20 episodes of care within their specialty specific s for Aetna members over the past three years. A reasonable volume of Aetna members is necessary to credibly measure performance. Symmetry Clean Period Each has its own clean period. It is defined as the absence of treatment for a specified period of time. For example, 204 (Conjunctivitis) has a 60-day clean period, which means that any claims related to that diagnosis that fall within a 60-day period will be considered a recurrence of the same condition. After 60 days, a new episode would be triggered. Only full-year or complete episodes are used for Aexcel. Complete episodes are those that met the clean periods before and after the measure, or episodes lasting 365 days. Aetna limits episodes to a maximum of 365 days. Claims Lag There is a claims lag of three months used in the episodes of care measurement. When episodes of care are used for Aexcel, we also apply a lag to allow for as many episodes to become complete or full-year episodes as possible for the measurement period. Episode Risk Groups Symmetry's Episode Risk Groups (ERGs ) version 6.5 is a software analysis tool that allows the use of enrollment information (demographics), medical claims, pharmacy claims and episodes of care in developing a risk-stratification score for individuals and groups. That score represents health care risk or burden of illness and can be applied to populations, such as a provider s patient base, to determine differences between the health burden of that provider versus his/her peers. ERGs can be retrospective, using historical claims to determine current risk, or prospective, to predict risk going forward. In Aexcel, we use retrospective ERG scores to determine the risk and burden of illness for each provider's Aetna practice eligible for Aexcel. That risk/burden of illness is then used during the development of normative values for case-mix adjustment to ensure that health risk and burden are accounted for in efficiency measurement when the provider is compared to his/her peers. Copyright Aetna Inc. 33

34 Aetna Case-Mix Adjustment The case-mix adjustment includes the following default variables for each individual used: Variable within each Aexcel Market Product Year of Episode Episode Risk Group (ERG) Specialty Category of the Provider Pharmacy Coverage Age Group Gender Values See Appendix C All medical products Calendar years 2004, 2005, 2006 and 2007, based on episode start date Risk group ranges from 1 to 4 where: Risk group 1: 0 to <0.7 Risk group 2: 0.7 to <2.4 Risk group 3: 2.4 to < 7.0 Risk group 4: 7.0 and greater See Appendix A Yes or No Age group ranges: Member ages 0 to 20 Member ages 21 to 50 Member ages 51 to 64 (Episodes with members age 65 and greater are excluded from the efficiency measurement.) Male or Female All variables are evaluated yearly. The items are then used in indirect standardization to create a grid of averages broken out by all of the above variables. If the cell is too small (less than 20), then the regional average allowed amount is used instead. The regional average allowed amount uses all the same variables with the exception of Region, which is substituted for Aexcel market and Specialty category of the provider group. For example, the following seven members would each fall into a separate case mix category and produce a different peer average for the case mix: Provider Specialty Aexcel Market Age/Age Group Gender Pharmacy Benefit Product Year of Episode ERG Average Allowed Amount* 268 Cardiology Atlanta 18 (0 to 20) Male Yes PPO $2, Cardiology Atlanta 25 (21-50) Male Yes PPO $ Cardiology Atlanta 60 (51-64) Male Yes PPO $2, Cardiology Atlanta 18 (0 to 20) Female Yes PPO $2, Cardiology Atlanta 25 (21-50) Female Yes PPO $2, Cardiology Atlanta 60 (51-64) Female Yes PPO $2, Cardiology Atlanta 18 (0 to 20) Male Yes HMO $1,800 *Average allowed amounts in this chart are for illustration purposes only. They do not represent actual average allowed amounts. In the table above, the difference between the first three case-mix categories is age. The fourth case-mix category yields a difference in age between case-mix categories two and three and a difference in gender from case-mix category one. Each of these groups would be considered a different case-mix category and would yield a different expected value. That goes for every variable combination that is encountered. Each combination of the case-mix variables produces its own expected value. This allows for a provider to be assessed while adjusting for the case-mix of the members he/she is treating, thus allowing for a more even Copyright Aetna Inc. 34

35 playing field to be used when comparing one provider to another and to the specialty/market average for each. Aetna Outlier Logic Outlier logic is applied at the case-mix category level as described above. When case-mix categories are created, the highest and lowest five percent of episodes within each case-mix category, based on the total cost, are considered outliers and, therefore, excluded. There is additional clinical logic applied to identify outliers that are based on the length of an episode. For example, an episode for an acute myocardial infarction that is only one day in length would be considered an outlier as it most likely represents a coding error and not an actual episode of acute myocardial infarction. Eligible Population Product Lines All medical products Administrative Specification Ages Members less than 65 Continuous Enrollment Benefit Provider Specialty Category Complete and full-year episodes are included, which means the episode began and came to an end according to the rules of clean periods or were 12-month episodes. Episodes longer than 12 months for chronic conditions are not created by Aetna s warehouse to enable comparisons. For example, if a chronic episode for one person is only 12 months long, it would not be comparable to an episode for another person with the same condition that is 3 years long. Medical (pharmacy if available) Cardiology, cardiothoracic surgery, gastroenterology, neurology, neurosurgery, vascular surgery, plastic surgery, general surgery, Ob/Gyn, orthopedic, urology, otolaryngology Measurement Timeframe Denominator Number of Episodes Three full years of episodes where the episode start begins within the three-year timeframe Denominator is the sum of the expected allowed amounts for the episodes being attributed to the group. The expected allowed amount is the case-mix-adjusted average for each created within each market/specialty combination The volume threshold for a provider/provider group to be considered for Aexcel is a minimum of 20 episodes. Total Allowed $ Episodes must be greater than $0 Numerator Numerator is the sum of the actual allowed amounts for the same episodes in the denominator being attributed to the group. Exclusion (1) All members age 65 and older (2) All transplants Copyright Aetna Inc. 35

36 Provider Grouping Provider data is sent to each region in the Aexcel Designation Model (see following section) for provider group selection. Provider groups can be either an individual or a group, based on how individual markets conduct business. Provider Scoring Scoring is done at the provider group level for volume. A provider group can include from one to many individual providers. Each provider group has the potential to be scored for efficiency, but it must meet the measure threshold of 20 or more episodes. Each group meeting the 20 or more threshold is considered for designation as long as it passes at least one of the clinical performance measures. Each group will receive an observed (actual rate) to expected (case-mixadjusted) ratio score. The expected value for this score is the case-mix-adjusted average allowed amount for each episode attributed to that group Statistical Significance of Efficiency Index Statistical significance is determined using a weighted Student s t-test in which we compare each provider group s episodes index to the mean episode index to test for statistical significance using a 90 percent confidence interval. Physician groups can be placed into four categories of statistical significance of efficiency: ESS - Efficient and statistically so ENSS - Efficient but not statistically so INSS Did not meet efficiency standards but not statistically so ISS Did not meet efficiency standards and statistically so NA - not enough volume to determine efficiency and statistical significance Copyright Aetna Inc. 36

37 Aexcel Designation Model Process The Aexcel designation process includes four key criteria: Case volume Clinical performance Efficiency Network adequacy For specialists who meet the case volume and clinical performance standards for Aexcel network designation, a measure of the efficiency of their care is developed and compared to that of their peers. Designation Process (performed by specialty in each market s designation model) 1. As described above, provider groups are assigned an efficiency/statistical significance type: a. ESS b. ENSS c. INSS d. ISS e. NA 2. The provider group s overall efficiency/statistical significance is percentiled within the market/specialty and stated as an add-back priority order to indicate relative performance within efficiency/statistical significance types 3. Provider groups are non-designated for Aexcel, regardless of efficiency/statistical significance type, if: a. They do not pass one of the clinical measures b. They are under investigation by the Special Investigations Unit 4. Provider groups are moved from a non-designated to a designated status, regardless of efficiency/statistical significance type, if they weren t in a non-designated status in the preceding step, and if: a. They have contractual language requiring them to be included in all networks b. They represent a rare sub-specialty or perform unique services within the market c. They are a multi-specialty group, and all of their specialties as a whole are deemed to meet the efficiency and general access standards 5. After designation/non-designation, the eligible network for each market and specialty category is built by automatically excluding all ISS and NA providers: however, additional providers can be added back, in add-back priority order, until the network size is sufficient to meet general access and penetration requirements (see note below regarding network adequacy). 6. A GeoAccess analysis is then conducted to determine if there are gaps in each market s geographic coverage using up to a 75-mile threshold. If gaps are identified, providers are added to the network, in add-back priority order, to address the geographic gaps. Copyright Aetna Inc. 37

38 Note: Aexcel Network Adequacy The Aexcel designation model produces a baseline Aexcel network based on the evaluation of volume, clinical performance, fraud flags and efficiency. Some physicians or physician groups also have Aexcel designation based on contractual obligations. Once these selections are complete, we may need to supplement the performance network with additional physicians to ensure members have satisfactory access to enough specialists. However, only physicians who have passed the clinical performance evaluation are eligible for consideration to supplement the network. Copyright Aetna Inc. 38

39 Appendix A: Aexcel Specialties and Sub Categories Specialty Sub Category Cardiology Cardiothoracic Surgery Gastroenterology Neurology Neurosurgery Obstetrics-Gynecology Orthopedics Otolaryngology Plastic Surgery Surgery Urology Vascular Surgery Cardiology Cardiovascular Disease Interventional Cardiology Surgery, Thoracic Cardiovascular Thoracic Surgery Gastroenterology Neurology Surgery, Neurological Obstetrics & Gynecology Gynecology Surgery, Obstetrics & Gynecology Obstetrics & Gynecology - CA PCP Surgery, Hand/Orthopedic Surgery, Knee Surgery, Orthopedic Otolaryngology Otology Otology/Neurotology Otorhinolaryngology Surgery, Head & Neck Surgery, Plastic Surgery, Hand/Plastic Surgery, Oro-Facial Plastic Surgery, Plastic and Reconstructive Otorhinolaryngology & Oro-Facial Plastic Surgery Otorhinolaryngology/Plastic Surgery Surgery Proctology Surgery, Colon & Rectal Surgery, Hand Urology Surgery, Urological Surgery, General Vascular Copyright Aetna Inc. 39

40 Appendix B: Specialty Groupings Code CARDIOLOGY Description CARDIOTHORACIC SURGERY Code Description 265 IHD, except CHF, w/o AMI 260 CAD, w/o AMI, w CABG 311 Cardiovascular signs/symptoms 251 CAD, w AMI, w CABG 277 Minor conduction disorder 400 Malig pulmonary neopl, w surg 281 Benign hypertension, w/o cc 900 Signs/symptoms non-specific 274 Valvular disorder, w/o cc 261 CAD/valve dis, w/o AMI w valve 280 Benign hypertension, w cc 290 Minor arterial inflam w/o surg 283 Cardiac congenital w/o surg 410 Pulmonary signs & symptoms 264 CAD w/o AMI, with cardiac cath 295 Atherosclerosis, w/o surgery 047 Hyperlipidemia 270 Aortic aneurysm, w/o surgery 279 Malignant hypertension, w/o cc 810 Late effect/late complication 262 CAD, w/o AMI, with angioplasty 401 Malig pulmonary neopl w/o surg 410 Pulmonary signs & symptoms 282 Cardiac congenital w surg 276 Maj conduction w/o pacer/defib 263 CAD w/o AMI w arrythmia and pacer 900 Signs/symptoms non-specific 373 Bacterial lung infection, w cc 268 Congestive Heart Failure w/o c 265 IHD, except CHF, w/o AMI 253 CAD, w AMI, w angioplasty 794 Routine exam 797 Conditional exam 399 Oth inflam lung dis w/o surg 794 Routine exam 292 Artery embolism/thromb w/o sur 278 Malignant hypertension, w cc 269 Aortic aneurysm, with surgery 267 Congestive Heart Failure, w cc 403 Benign pulmonary neopl w/o sur 029 Non-insulin diabetes w cc 159 CVA, non-hemorrhagic w surgery 160 CVA, non-hemorrhagic, w/o surg 310 Other diseases of the veins 022 Hypo-function thyroid gland 678 Minor inflam skin/subcu tissue 048 Obesity, mild 305 Phlebitis/thrombophleb vein 290 Minor arterial inflam w/o surg 302 Vein inflammation, w surgery 255 CAD, w AMI, w cardiac cath 271 Cardiac infection, w surgery 287 Other cardiac diseases 797 Conditional exam 295 Atherosclerosis, w/o surgery 677 Major inflam skin/subcu tissue 272 Cardiac infection, w/o surgery 288 Arterial inflammation w surg 259 CAD w AMI inferior wall w/o cc 374 Bacterial lung infect w/o cc 263 CAD w/o AMI w arrythmia and pacer 294 Atherosclerosis, with surgery 027 Insulin depend diabetes w cc 402 Benign pulmonary neopl w surg 266 Pulmonary heart disease w/oami 252 CAD w AMI/aqua defect, w valve 257 CAD w AMI anterior wall w/o cc 258 CAD, w AMI inferior wall, w cc 270 Aortic aneurysm, w/o surgery 030 Non-insulin diabetes w/o cc 273 Valvular disorder, w cc 303 Embolism/thrombosis veins 305 Phlebitis/thrombophleb vein 256 CAD, w AMI anterior wall, w cc 275 Major conduction w pacer/defib 028 Insulin depend diabetes w/o cc 254 CAD w AMI, w arrythmia w pacer Copyright Aetna Inc. 40

41 GASTROENTEROLOGY NEUROLOGY Code Description Code Description 433 Inflam esophagus, w/o surgery 168 Migraine headache, common 449 Diverticulitis, w/o surgery 900 Signs/symptoms non-specific 486 Gastroenterology signs&symptom 185 Neurological signs & symptoms 464 Irritable bowel syndrome 722 Joint degen localized, w/o sur 455 Benign neo intest/abdom w surg 177 Inflam non-cran nerves w/o sur 474 Hemorrhoids, simple 169 Migraine headache, complicated 794 Routine exam 171 Congenital/other CNS w/o surg 435 Gastritis/duodenitis, simple 160 CVA, non-hemorrhagic, w/o surg 456 Benign neo intest/abd w/o surg 152 Epilepsy, w/o surgery 452 Inflam intestine/abd w/o sur 175 Carpal tunnel syndrome w/o sur 485 Oth dis rectum/anus w/o surg 167 Hereditary/degener CNS w/o sur 479 Benign neo rectum/anus, w surg 150 Inflammation CNS w/o surg 438 Ulcer, simple 173 Inflam cranial nerves w/o surg 076 Non-neopl blood disease, minor 101 Attention deficit disorder 473 Hemorrhoids, comp, w/o surgery 184 Congenital peripheral nerve 900 Signs/symptoms non-specific 749 Oth minor orthopedic disorder 533 Hepatology signs & symptoms 353 Other ENT disorder w/o surgery 431 Infect stomach/esophag w/o cc 106 Other neuropsych/behavior dis 469 Oth diseases intestine/abdomen 746 Minor orthopedic trauma 468 Hiatal hernia, w/o surgery 102 Development disorder 514 Infect hepatitis L sev w/o cc 163 Minor brain trauma 484 Oth disorder rectum/anus w sur 752 Ortho/rheumatoid signs&symptom 442 Benign neo stoma/esoph w/o sur 092 Autism and child psychoses 450 Oth infect intestines/abdomen 091 Organic drug/metabolic dis 480 Benign neo rectum/anus w/o sur 206 Inflam eye disease w/o surgery 518 Cirrhosis, w/o surgery 156 Benign neoplasm CNS w/o surg 437 Ulcer, complicated w/o surgery 183 Trauma non-cran nerve w/o surg 521 Cholelithiasis, complicated 751 Ortho cong/acquir deform w/o s 445 Anom stomach/esophagus w/o sur 165 Spinal trauma, w/o surgery 463 Bowel obstruction, w/o surgery 158 CVA, hemorrhagic, w/o surgery 519 Acute pancreatitis 154 Malig neoplasm CNS w/o surg 516 Non-infect hepatitis, w/o cc 140 Viral meningitis 523 Cholelithiasis simple w/o surg 162 Major brain trauma, w/o surg 520 Chronic pancreatitis 179 Peripheral nerve neo w/o surg 512 Infect hepatitis H sev w/o cc 143 Nonviral encephalitis 513 Infect hepatitis, low sev w cc 141 Bacterial/fungal meningitis 142 Viral encephalitis 181 Trauma cranial nerves w/o surg 145 Toxic encephalitis Copyright Aetna Inc. 41

42 NEUROSURGERY OB/GYN Code Description Code Description 722 Joint degen localized, w/o sur 794 Routine exam 721 Joint degen localized, w surg 649 Cond asso w menstruation w/o s 156 Benign neoplasm CNS w/o surg 640 Infect vagina except monilial 171 Congenital/other CNS w/o surg 647 Benign neo female genit w/o s 749 Oth minor orthopedic disorder 661 Gynecological signs & symptoms 751 Ortho cong/acquir deform w/o s 574 Infect low GU not STD 153 Malignant neoplasm CNS w surg 658 Benign neo breast, w/o surgery 746 Minor orthopedic trauma 639 Monilial infect vagina (yeast) 177 Inflam non-cran nerves w/o sur 653 Oth female genital cond w/o s 160 CVA, non-hemorrhagic, w/o surg 643 Inflam female genital w/o surg 162 Major brain trauma, w/o surg 796 Contraceptive manage, w/o surg 161 Major brain trauma, w surgery 623 Induced abortion 165 Spinal trauma, w/o surgery 646 Benign neo female genital w s 810 Late effect/late complication 622 Spontaneous abortion 167 Hereditary/degener CNS w/o sur 637 Infection cervix, w/o surgery 034 Benign neoplasm pituitary 572 Sexually trans disease primary 166 Hereditary/degener CNS w surg 786 Oth maj neonatal, perinatal 175 Carpal tunnel syndrome w/o sur 780 Uncomp neonatal management 173 Inflam cranial nerves w/o surg 797 Conditional exam 174 Carpal tunnel syndrome w surg 592 Urinary incontinence w/o surg 163 Minor brain trauma 720 Osteoporosis 158 CVA, hemorrhagic, w/o surgery 651 Cond asso w infertility w/o s 155 Benign neoplasm CNS w surg 648 Cond asso w menstruation w sur 185 Neurological signs & symptoms 790 Exposure to infectious disease 154 Malig neoplasm CNS w/o surg 645 Malig female genital w/o surg 170 Congenital/other CNS w surgery 596 Urologic signs & symptoms 164 Spinal trauma, with surgery 801 Other preventative/adm service 159 CVA, non-hemorrhagic w surgery 795 Contraceptive manage, w surg 632 Infect ovary/tube w/o s w/o cc 642 Endometriosis, w/o surgery 580 Inflam genitourinary w/o surg 787 Oth minor neonatal, perinatal 652 Oth female genital cond w surg 595 Oth genitourinary sys w/o surg 650 Cond asso w infertility w surg 784 Mechanical disorder antenatal 638 Vaginal infection, w surgery 641 Inflam female genital, w surg 621 Ectopic pregnancy, w/o surgery 630 Infect ovary/fallop tube w sur 591 Urinary incontinence w surgery 620 Ectopic pregnancy, w surgery 644 Malig female genital w surg 635 Infect uterus w/o surg w/o cc 588 Benign neo GU w/o surg 573 Sexually trans disease dissem 785 Other disorders, antenatal 631 Infect ovary/tube w/o sur w cc 633 Infection uterus, with surgery 634 Infection uterus w/o surg w cc Copyright Aetna Inc. 42

43 ORTHOPEDICS OTOLARYNGOLOGY Code Description Code Description 722 Joint degen localized, w/o sur 329 Otitis media, w/o surgery 746 Minor orthopedic trauma 335 Chronic sinusitis, w/o surgery 748 Bursitis/tendinitis, w/o surg 331 Tonsill/adenoid/pharyn w/o sur 743 Joint derangement, w/o surgery 341 Minor ENT inflam cond w/o surg 731 Cl frac/dis up extrem w/o surg 337 Oth ENT infection w/o surgery 751 Ortho cong/acquir deform w/o s 349 Hearing disorders, w/o surgery 749 Oth minor orthopedic disorder 330 Tonsill/adenoid/pharyn w surg 752 Ortho/rheumatoid signs&symptom 354 Otolaryngology signs/symptoms 742 Joint derangement, w surgery 328 Otitis media, w minor surgery 734 Cl frac/dis low extrem w/o sur 339 Major ENT inflam cond w/o surg 721 Joint degen localized, w surg 353 Other ENT disorder w/o surgery 690 Minor skin trauma 900 Signs/symptoms non-specific 719 Major joint inflam, w/o surg 333 Acute sinusitis 175 Carpal tunnel syndrome w/o sur 347 ENT congenital/anomal w/o sur 733 Cl frac/dis low extrem w surg 332 Allergic rhinitis 745 Maj orth trauma except frac/dis w/o s 351 ENT trauma, w/o surgery 177 Inflam non-cran nerves w/o sur 433 Inflam esophagus, w/o surgery 730 Cl frac/dis up extrem w surg 334 Chronic sinusitis with surgery 174 Carpal tunnel syndrome w surg 345 Benign ENT neoplasm, w/o surg 747 Bursitis/tendinitis, w surg 185 Neurological signs & symptoms 741 Benign bone/conn ex head/neck 322 Inflam oral cavity w/o surgery 737 Cl fracture/dis trunk w/o surg 486 Gastroenterology signs&symptom 744 Maj ortho trauma ex frac/dis w s 682 Benign neoplasm of the skin 729 Op frac/dis of upper extremity 678 Minor inflam skin/subcu tissue 728 Cl frac/dis w/o s thi/hip/pelv 327 Otitis media, w major surgery 750 Ortho cong/acquir deform w sur 320 Infection of the oral cavity 727 Cl frac/dis w s thigh/hip/pelv 343 Malignant ENT neoplasm w/o sur 732 Op frac/dis lower extremity 346 ENT congenital/anomalies w sur 718 Major joint inflam w surg 338 Major ENT inflam cond w surg 736 Cl fracture/dis trunk w surg 671 Maj bact infection skin, w/o s 350 ENT trauma, with surgery Benign ENT neoplasm, w 344 surgery 672 Minor bacterial infection skin 336 Oth ENT infection with surgery 340 Minor ENT inflam cond w surg 679 Malig skin, major, w surgery 675 Fungal skin infection, w/o sur 342 Malignant ENT neoplasm w surg 352 Other ENT disorders, w surgery 348 Hearing disorders with surgery Copyright Aetna Inc. 43

44 PLASTIC SURGERY Code Description 682 Benign neoplasm of the skin 678 Minor inflam skin/subcu tissue 658 Benign neo breast, w/o surgery 679 Malig skin, major, w surgery 685 Major skin trauma ex burns w s 688 Open wound, with surgery 657 Benign neo breast, w surgery 689 Open wound, w/o surgery 900 Signs/symptoms non-specific 749 Oth minor orthopedic disorder 686 Maj skin trauma ex burns w/o s 655 Malig breast w surgery ex BMT 743 Joint derangement, w/o surgery 672 Minor bacterial infection skin 810 Late effect/late complication 175 Carpal tunnel syndrome w/o sur 741 Benign bone/conn ex head/neck 731 Cl frac/dis up extrem w/o surg 680 Malig skin, major, w/o surgery 722 Joint degen localized, w/o sur 347 ENT congenital/anomal w/o sur 677 Major inflam skin/subcu tissue 174 Carpal tunnel syndrome w surg 748 Bursitis/tendinitis, w/o surg 690 Minor skin trauma 660 Oth disorders breast, w/o surg 751 Ortho cong/acquir deform w/o s 752 Ortho/rheumatoid signs&symptom 351 ENT trauma, w/o surgery 746 Minor orthopedic trauma 466 Hernias except hiatal w/o surg 687 Minor burns 346 ENT congenital/anomalies w sur 209 Malig neoplasm eye external 218 Trauma of the eye, w surgery 740 Benign bone/connect head/neck 798 Major specific procedure NOS 691 Other skin disorders 354 Otolaryngology signs/symptoms 345 Benign ENT neoplasm, w/o surg 341 Minor ENT inflam cond w/o surg 306 Varicose veins lower extremity 343 Malignant ENT neoplasm w/o sur 349 Hearing disorders, w/o surgery Copyright Aetna Inc. 44

45 SURGERY Code Description Code Description 658 Benign neo breast, w/o surgery 523 Cholelithiasis simple w/o surg 465 Hernias, except hiatal, w surg 689 Open wound, w/o surgery 678 Minor inflam skin/subcu tissue 677 Major inflam skin/subcu tissue 473 Hemorrhoids, comp, w/o surgery 749 Oth minor orthopedic disorder 466 Hernias except hiatal w/o surg 479 Benign neo rectum/anus, w surg 682 Benign neoplasm of the skin 690 Minor skin trauma 657 Benign neo breast, w surgery 446 Appendicitis, with rupture 661 Gynecological signs & symptoms 452 Inflam intestine/abd w/o sur 521 Cholelithiasis, complicated 477 Malig rectum/anus, w surgery 486 Gastroenterology signs&symptom 451 Inflam intestine/abdomen w sur 900 Signs/symptoms non-specific 463 Bowel obstruction, w/o surgery 794 Routine exam 470 Infection rectum/anus, w surg 672 Minor bacterial infection skin 679 Malig skin, major, w surgery 449 Diverticulitis, w/o surgery 302 Vein inflammation, w surgery 522 Cholelithiasis, simple, w surg 049 Obesity, morbid with surgery 447 Appendicitis, w/o rupture 462 Bowel obstruction with surgery 474 Hemorrhoids, simple 456 Benign neo intest/abd w/o surg 655 Malig breast w surgery ex BMT 484 Oth disorder rectum/anus w sur 476 Inflam rectum/anus w/o surgery 448 Diverticulitis, with surgery 306 Varicose veins lower extremity 433 Inflam esophagus, w/o surgery 050 Obesity, morbid w/o surgery 453 Malig intestine/abdomen w surg 455 Benign neo intest/abdom w surg 020 Disease-thyroid gland w surg 656 Malignancy breast, w/o surgery 475 Inflam rectum/anus, w surgery 472 Hemorrhoids, comp, w surg 480 Benign neo rectum/anus w/o sur 485 Oth dis rectum/anus w/o surg 159 CVA, non-hemorrhagic w surgery 660 Oth disorders breast, w/o surg 478 Malig rectum/anus, w/o surgery 290 Minor arterial inflam w/o surg 454 Malig intestine/abd w/o surg 741 Benign bone/conn ex head/neck 685 Major skin trauma ex burns w s 810 Late effect/late complication 461 Vascular disease intest/abdom 295 Atherosclerosis, w/o surgery Copyright Aetna Inc. 45

46 UROLOGY VASCULAR SURGERY Code Description Code Description 574 Infect low GU, not STD 465 Hernias, except hiatal, w surg 580 Inflam genitourinary w/o surg 305 Phlebitis/thrombophleb vein 596 Urologic signs & symptoms 658 Benign neo breast, w/o surgery 584 Benign neo prostate, w/o surg 302 Vein inflammation, w surgery 578 Kidney stone w/o surg & comorb 682 Benign neoplasm of the skin 592 Urinary incontinence w/o surg 752 Ortho/rheumatoid signs&symptom 582 Malignancy prostate, w/o surg 159 CVA, non-hemorrhagic w surgery 595 Oth genitourinary sys w/o surg 678 Minor inflam skin/subcu tissue 576 Kidney stone w surg w/o comorb 466 Hernias except hiatal w/o surg 795 Contraceptive manage, w surg 553 Chronic renal failure, w ESRD 044 Male sex gland disorders 294 Atherosclerosis, with surgery 586 Malig genitourinary w/o surg 292 Artery embolism/thromb w/o sur 794 Routine exam 521 Cholelithiasis, complicated 307 Oth minor inflam disease veins 677 Major inflam skin/subcu tissue 593 Male Infertility 672 Minor bacterial infection skin 588 Benign neo GU w/o surg 900 Signs/symptoms non-specific 673 Viral skin infection 303 Embolism/thrombosis veins 585 Malignancy genitourinary w sur 269 Aortic aneurysm, with surgery 579 Inflam genitourinary w surgery 657 Benign neo breast, w surgery 583 Benign neo prostate, w surgery 297 Art aneurysm ex aorta w/o surg 581 Malignancy prostate, w surgery 304 Disorder of lymphatic channels 594 Oth genitourinary with surgery 050 Obesity, morbid w/o surgery 653 Oth female genital cond w/o s 486 Gastroenterology signs&symptom 591 Urinary incontinence w surgery 722 Joint degen localized, w/o sur 571 Infect upper GU w/o s 810 Late effect/late complication 577 Kidney stone w/o sur w comorb 522 Cholelithiasis, simple, w surg 575 Kidney stone w surg, w comorb 177 Inflam non-cran nerves w/o sur 291 Maj non-inflam artery dis w s 447 Appendicitis, w/o rupture 749 Oth minor orthopedic disorder 029 Non-insulin diabetes w cc 554 Chronic renal failure w/o ESRD 288 Arterial inflammation w surg 296 Artery aneurysm ex aorta w sur Copyright Aetna Inc. 46

47 Appendix C 2009 Aexcel Markets Arizona Atlanta, GA Austin, TX Central Valley, CA Chicago, IL Cincinnati, OH Cleveland, OH Colorado Columbus, OH Connecticut Dallas/Fort Worth, TX Delaware Detroit, MI Houston, TX Indianapolis, IN Kansas City (KS and MO) Los Angeles, CA Louisville, KY Maine Metropolitan DC (including MD, DC and N. VA) Metropolitan New York North Florida Northern CA Northern New Jersey Oklahoma City, OK Orlando, FL Pittsburgh, PA Richmond, VA San Antonio, TX San Diego, CA Seattle/W. Washington South FL (Dade and Broward Counties) Tampa, FL Tulsa, OK Copyright Aetna Inc. 47

48 A CONDITION CLASSIFICATION AND EPISODE BUILDING SYSTEM Introduction A condition classification system which identifies discretely occurring episodes of care has long been at the center of analytical demand for those health care professionals responsible for the funding, analysis and direct provision of health care services. In simple terms, unless one can consistently and meaningfully quantify those services offered by health care providers, attempts to manage supply and demand or otherwise to effect change is elusive at best. This desire, then, for a meaningful unit of analysis lies at the heart of Symmetry's efforts with respect to its Episode Treatment Groups (s). Previously, work in this area relied solely on the identification of discrete patient groups in an inpatient setting (1) or the simple "bucketing" of diagnosis codes (2). Diagnosis Related Groups (DRGs) provided an easily understood classification system which organized inpatient stays into one of approximately 500 groups. The DRG methodology was developed in the late 1970's at Yale University and subsequently adopted by Medicare as a prospective payment methodology. In short, this classification scheme provided a means to identify and measure inpatient confinements, and further, formed the basis from which to identify and compare resource consumption in terms of dollars. DRGs as a unit of analysis have been widely used in the private sector by both purchasers and providers of health care services to quantify both demand and relative financial performance. Its wide acceptance as a tool to define an inpatient population is due principally to the methodology underlying DRGs. This classification scheme proved a reliable case mix adjustment method because it controlled for patient severity by establishing a manageable number of clinically homogeneous groups. This characteristic coupled with the statistical stability of its groups has made DRGs the most widely used patient classification methodology. Unfortunately, DRGs do not address ambulatory data, nor do they create a unit of analysis based on a complete episode of care spanning both inpatient and outpatient settings. Diagnosis Clusters, originally developed by Schneeweiss and his colleagues at the University of Washington for use in the ambulatory setting, essentially groups diagnosis codes together into similar "clusters". Diagnosis Clusters were not developed as a methodology to adjust for case mix, to account for resource use or to establish a treatment episode. Other classification methods have been developed as a basis for reimbursement (3), financing (4), or as a means to control for case mix but without benefit of creating a clinically homogenous unit or an episode of care (5). s The methodology is similar to that of the DRGs but with several important differences. Perhaps the most obvious is that s identify and classify an entire episode of care regardless of whether the patient has received medical treatment as an outpatient, inpatient, or both. Specifically, the characteristics of the methodology are: Case mix adjustment s have been developed to account for differences in patient severity. Differences in patient age, complicating conditions, comorbidities and major surgeries have been factored into the definition of the s. Clinical Homogeneity Each is clinically homogeneous with respect to the underlying condition, and hence treatment requirements.

49 Episode Building s build a complete treatment episode which incorporates both inpatient and ambulatory care, including pharmaceutical services. Once treatment for an episode has begun, the software continues to collect all clinically relevant information until an absence of treatment or 'clean-period' is detected. This insures that all appropriate procedural and cost information has been collected and correctly assigned to one complete treatment episode. Concurrent and Recurrent Episodes Using the service unit of an individual claim or encounter form as input, s identify and track the treatment of different conditions which can occur even during the same patient encounter. As a result, s separate and identify concurrently occurring conditions and assigns each health care service to the clinically appropriate episode. In addition, should a patient be successfully treated but suffer a recurrence of the same condition, the software identifies recurrent episodes. Shifting Episodes s account for changes in a patient's condition during the course of treatment. Once a change in condition has been identified, the patient's entire episode shifts from the initially defined to the which includes the change in condition. In this way, the progression of a condition is identified. Manageable Number of Groups The episode-building condition classification system is comprised of 574 clinically homogeneous and statistically stable groups. A complete list of the s, their English descriptions and corresponding Major Practice Categories (MPCs) can be found here. General Philosophy s are designed to provide a consistent and reliable measurement tool to measure the provision and financing of health care services. Specifically, the s can serve as: An analytical unit in which to measure and compare the utilization and financial performance of health care providers (provider profiling) A clinically useful unit from which to measure health care demand A basis to establish disease management strategies, especially with the inclusion of pharmaceutical claims As a point of reference, s were not primarily designed to: Detect inappropriate diagnosis/procedure code combinations for the purposes of clinical or payment review or denial Identify potentially inappropriate health care services Provide the basis for a population rating mechanism Be used as an encounter-based reimbursement methodology Case Mix Adjustment and Clinical Homogeneity At the core of the methodology are two concepts which are interrelated: case mix adjustment and clinical homogeneity. In short, case mix adjustment is a term used to describe the goal of a classification scheme which endeavors to explain, among other things, resource consumption. Although other terms like "resource intensity" or "severity" are used, both are proxy measures for the type and amount of resources which classification methods hope to explain. This is usually accomplished by identifying discrete units of patients or conditions which differ from one another with respect to resource consumption. Once these discrete groups of patients are identified, any subsequent analysis based upon these discrete units can be said to be "case mix adjusted". Of course the ability to use these discrete units varies widely with the method used to define them. In general, the wider the definition of each group, the greater the variability of resource consumption among the members of the group. And the greater the variability, the weaker the method's ability to distinguish real differences between and among populations. Towards the other end of the spectrum, as the definition of each group narrows, the explanation of resource consumption increases. At its most extreme, each patient or each patient's condition becomes its own group. The result is a perfect explanation of resource consumption but since each unit is absolutely unique with respect to the others, comparisons become impossible. A balance exists between a manageable number of uniquely defined groups and the explanation of resource consumption.

50 Clinical homogeneity is a term used to describe the extent to which those patients or patient's conditions are similar from a clinical perspective. Patients or conditions which are clinically similar provide a natural basis (especially among clinicians) from which to identify differences among populations. A clinically homogeneous group of asthma cases, for example, provides a more intuitive unit of analysis than a group of patients with dis-similar diagnostic characteristics. Additionally, while there is a natural tendency to simply place all patients with the same diagnosis in a single group, there are often other characteristics which may further differentiate those patients with respect to their severity. Asthma patients, for example, generally differ in their level of illness depending upon their age and whether there exists a specific comorbid condition. Hence, adjusting for the patient's age and comorbidity improves clinical homogeneity. s adjust for case mix, initially, by grouping together clinically similar conditions; for example, the several diagnosis codes which together are referred to as chronic bronchitis. Further differentiation of patient episodes may occur based on clinical complications, the existence of complicating comorbidities, patient age, the chronicity or severity of the condition, the significant use of surgery as treatment, and the level of management. Each is described below. Complications Complications which develop during the evolution of an episode can influence the amount of resources used to treat the condition. In chronic bronchitis for example, a patient who develops pneumonia will incur higher cost than a patient without pneumonia. Therefore, the grouper software makes a distinction between these two types of chronic bronchitis, i.e., chronic bronchitis with complication and Chronic Bronchitis without complication. Complications are specific: a complication of pneumonia for chronic bronchitis does not automatically affect another concurrent episode with a complication distinction. Comorbidities In some cases, a patient's prior medical history may impact the type and volume of resources required to treat a condition episode. Again for chronic bronchitis, a patient with a history of asthma will generally require additional resources than a patient without asthma. Hence, the methodology specifically identifies chronic bronchitis with complication with comorbidity from chronic bronchitis with complication without comorbidity. Rare and Chronic Conditions Rare conditions generally are included in a broader illness category to avoid the problem of having too few members in a group. However certain conditions such as blood clotting disorders require such intensive and expensive management that they occupy their own unique group. Chronic conditions such as insulin dependent diabetes must be considered differently from conditions that are expected to have an onset and a completion of care. These conditions occupy unique groups, and can be managed in yearly units as opposed to using an episode as the unit of analysis. Age In other cases, the age of a patient can affect the extent of resource utilization for the treatment of a particular condition. Therefore, some s are differentiated based on the age of an individual at the onset of an episode, hence: Acute bronchitis with comorbidity age less than 5. Defining Surgeries In those instances in which a particular surgery is indicative of a 'sicker' patient, s account for the distinct differences in resource consumption between those patients with the surgical procedure from those without. The classification of a patient's condition into the with surgery does not occur unless a specific surgical procedure(s), called a defining surgery, has been provided. As a general rule, defining surgeries have the following characteristic: Defining surgeries do not include purely diagnostic procedures or biopsies since they do not necessarily identify a 'sicker' patient. For example, Diagnostic Esophagoscopy (CPT-4: 43200) is not a defining surgery for Inflammation of the esophagus, with surgery. However, Esophagoscopy with the removal of tumor(s), polyp(s) or other lesion(s) (CPT-4: 43216) is a defining surgery.

51 These defining surgeries are specific to an : While a hip replacement procedure is a defining surgery for the treatment of a patient's Hip Fracture, the surgery has no effect on any other currently occurring episodes for the patient, e.g., Acute Bronchitis. Non-Active Management vs. Active Management Cancer is a condition where treatment decisions are based upon myriad clinical and patient choice issues. Treatments vary from watchful waiting to aggressive management to post-treatment observation. A decision to manage a cancer aggressively results in a significant increase in expected resource consumption. Therefore cancers are divided into those episodes which are being actively managed by radiation and/or chemotherapy, and those which are not.

52 Structure of Episode Treatment Groups Similar to the structure of DRGs, whereby each DRG belongs to one Major Diagnostic Category defined by body system, each belongs to a specific Major Practice Category (MPC). Each MPC represents a body system and/or a particular physician specialty. In the flowchart example in Figure 1, a small portion of the Gastroenterology MPC is displayed. Note that s 430 and 431 correspond to gastroesophageal infection, with and without comorbidity, respectively. Inflammation of the gastroesophageal tract is further identified by site; esophagus and stomach/duodenum. Inflammation of the esophagus is further sub-defined by defining surgery. Each of the 22 MPCs are similarly arranged. Figure 1 Gastroesophageal Section of MPC 11: Gastroenterology Clinical Logic and Shifting The basis of the clinical logic for the methodology are a series of Diagnosis and Procedure code "eligibility" tables. Each and every ICD-9 (diagnosis), CPT-4 (procedure), NDC (drug) and HCPCS (procedure) code has been laboriously mapped to each of the s. As each claim record is considered by the grouper, the procedure and diagnosis codes are evaluated with respect to each other and in turn to each of the eligibility tables. Those "matches" are then considered with respect to the s. Diagnosis Codes Every diagnosis code is accounted for by the methodology, with the exception of the 'E' codes. With respect to the diagnosis code eligibility table, each ICD-9 code is principally assigned to one and only one. This mapping forms the basis for initial assignment; Initial, since subsequent diagnosis codes may affect final assignment in a process called shifting. The way in which a diagnosis code shifts the assignment of an episode depends on the class of ICD-9 code with respect to the s. A brief description of each class of diagnosis code follows. Primary diagnosis codes are the principal means of establishing initial assignment. Each diagnosis code is mapped as primary to one and only one. Incidental diagnoses are those codes which represent an illness or condition that is present during the treatment of another related, but usually more serious illness or condition. Incidental diagnoses do not shift an assignment.

53 For example, if during the course of treatment for acute bronchitis a patient is treated for "throat pain", the throat pain diagnosis is considered incidental to acute bronchitis. Rather than begin a new episode with throat pain, this claim record and the information it contains will be considered part of the acute bronchitis episode. From a chronological perspective, if a throat pain claim was considered first, it would still group to the acute bronchitis episode. Comorbid diagnoses represent an on-going chronic condition which affects a patient's treatment for a given condition episode. Diagnosis codes in this class will shift an assignment regardless of its chronological relationship to the episode. Only certain comorbid diagnoses will cause the to shift if it is identified prior to, concurrent with, or subsequent to the treatment of the episode. For example, a patient receiving care for arthritis would be considered to be more difficult and expensive to treat if the patient also had osteoporosis, a comorbid condition. Although the patient may not be concurrently receiving treatment for osteoporosis per se, the patient would generally require a more intensive treatment regimen for the arthritis than if this cormorbid condition did not exist. Hence, the patient shifts into a more complicated episode of arthritis with comorbidity. In addition, the software keeps track of each patient's comorbid conditions in a reference table. Therefore, a patient identified with certain comorbid conditions need not display a recent claim for, in this case, osteoporosis to shift into a more complicated. Comorbidities are specific. For example, osteoporosis would not be considered a comorbid diagnosis for an otolaryngology. Complicating diagnoses indicate that the patient treated for a current condition episode has developed a complication requiring a more intensive treatment protocol. Diagnosis codes in this class will shift an assignment only when the complication is identified subsequent to the start of the episode. For example, should a patient receiving treatment for bronchitis develop pneumonia, a known complication of bronchitis and more severe condition, the entire condition episode is defined as an episode of bronchitis with complication. As with comorbidities, complications are specific. Procedure Codes The procedure code eligibility table helps to identify the s in which a particular claim record can be assigned. Only those CPT-4 defined services which, from a clinical perspective, could conceivably group to a particular will do so. For example, if both a chest x-ray and blood glucose test were provided to a patient during the same encounter and further, if the patient had active episodes of both chronic bronchitis and diabetes; the chest x-ray will be assigned to the chronic bronchitis episode while the blood glucose test is assigned to the diabetes episode. In other words, the blood glucose test is not eligible for assignment to the chronic bronchitis. Defining Surgeries Similar to those comorbid and complicating diagnoses codes which affect the final assignment (shift) of an episode, there are specific surgeries which also affect an episode's final assignment. These are referred to as defining surgeries. As with the comorbid and complicating diagnoses, defining surgeries are specific and whose use is indicative of conditions with higher severity. For example, a patient receiving care for eye trauma requiring surgery to repair a detached retina is considered more severe and requires more intensive (and expensive) treatment than a patient who did not require detached retina repair. Pharmaceutical Claims As pharmaceutical claims typically do not contain diagnosis codes, the methodology includes the ability to identify the disease state for which a particular drug has been prescribed and assign pharmaceutical claims data to the most clinically appropriate episode.

54 National Drug Code National Drug Code (NDC) is an eleven digit code usually available on pharmaceutical claims, particularly when a Pharmacy Benefit Manager (PBM) plays some role in the management or adjudication of prescription drug claims. The NDC code contains highly specific information regarding each drug, its manufacturer and dosage. Using a sophisticated hierarchical approach, the methodology evaluates each prescription drug claim against each of the concurrently occurring episodes for which the patient is treated. A clinical logic base identifies for which of the concurrently occurring episodes a particular drug can be prescribed and then assigns the drug claim to the most clinically appropriate episode. Episode Creation Along with the clinical aspects, the method of episode creation is a crucial feature of the methodology. The approach taken for the identification of a complete episode relies on a flexible, rather than a fixed length of time. In other words, there are no presumed definitions of an episode's chronological length. The methodology will continue to identify and track all clinical activity for an episode for as long as a condition is actively treated. s accomplish this by the identification of discrete 'clean periods'. A clean period is defined as the absence of treatment for a specified period of time. Each has its own clean period. For example, the clean period for Acute bronchitis with comorbidity, age less than 5 is 60 days. Once an episode has begun for this, all clinically consistent claims activity for acute bronchitis will group to this episode until such time as 60 days passes without any corresponding clinically consistent treatment. For Chronic Bronchitis, with comorbidity on the other hand, the clean period is 180 days, consistent with an illness of greater and more chronic severity. In some obvious instances, e.g. benign hypertension, there is no clean period. The condition is basically lifelong and all clinically consistent treatment will group to an episode of benign hypertension for as long as data are available. This presents the analyst with some interesting challenges which will be discussed later in the section titled Complete and Incomplete Episodes. Assignment Thus far, the clinical and episode building characteristics of Episode Treatment Groups have been explained. The discussion which follows explains actual assignment and the methods employed by the methodology to build an episode. Expense Line Grouping A medical claim coming from a provider typically has two parts: a claim 'header', which contains patient and provider information as well as ICD-9 diagnoses; and the billing detail which contains service date information as well as billing procedure codes (usually CPT) and billing amounts. Upon data entry into a typical claims/encounter system, each claim will spawn as many 'expense line' records as there are billable procedures. Each expense line will contain all of the information from the 'header' and therefore, all diagnoses (maximum of four) appear on each expense line record. An example follows: Table 1 HEADER FILE Patient ICD-9 ICD-9 ICD-9 ICD-9 John Doe Service Date From 490 Bronchitis Diabetes Table 2 BILLING DETAIL Blank Blank Service Date To CPT-4 AMOUNT DESCRIPTION 07/17/96 07/17/ $X Office Visit 07/17/96 07/17/ $Y Chest x-ray 07/17/96 07/17/ $Z Blood Glucose

55 Patient ICD-9 ICD-9 ICD-9 ICD-9 John Doe John Doe John Doe 490 Bronchitis 490 Bronchitis 490 Bronchitis Diabetes Diabetes Diabetes Table 3 EXPENSE LINES Service Date From Service Date To CPT-4 AMOUNT DESCRIPTION Blank Blank 07/17/96 07/17/ $X Office Visit Blank Blank 07/17/96 07/17/ $Y Chest x-ray Blank Blank 07/17/96 07/17/ $Z Blood Glucose The grouping software evaluates each expense line record independently and determines which procedure code 'links' to which diagnosis code. Each diagnosis code on every expense line is evaluated with respect to the billed procedure code to determine eligibility. During this evaluation, one assumption is made: the first diagnosis code is considered primary. In the example above, both the bronchitis and diabetes diagnosis codes are evaluated with respect to each of the three procedure codes to determine 'matches' with respect to the s. When a match occurs, the expense line becomes a part of an episode corresponding to a particular. As with our example, the first expense line considered shows the two diagnoses (bronchitis and diabetes) and the billing procedure (office visit). While the office visit is considered valid for every, bronchitis is in the first diagnosis position and therefore this record will group to one of the four acute bronchitis s. (The specific bronchitis will be determined by a combination of the patient's age and the whether or not the patient has a comorbidity, i.e., asthma). The second expense line again shows the two diagnoses but the billing procedure is for a chest x-ray. In this case, the chest x-ray matches a diagnosis of bronchitis but does not match with the diagnosis of diabetes and therefore links to the bronchitis episode. The third expense line, of course, has both diagnoses but the blood glucose exam as the billing procedure. Since the blood glucose exam is inconsistent with a diagnosis of bronchitis but is consistent with diagnosis of diabetes, this expense line will link itself to a diabetes episode. Shown below is our three expense line example, following the clinical matching of each CPT procedure code to the appropriate diagnosis code. Table 4 EXPENSE LINES FOLLOWING CLINICAL MATCHING Patient ICD-9 ICD-9 ICD-9 ICD-9 CPT-4 AMOUNT DESCRIPTION EPISODE John Doe John Doe John Doe 490 Bronchitis 490 Bronchitis 490 Bronchitis Diabetes Diabetes Diabetes Blank Blank $X Office Visit Bronchitis Blank Blank $Y Chest x-ray Bronchitis Blank Blank $Z Blood Glucose Diabetes Reviewing this example, this one health care encounter has 'spawned' three separate billable expense lines which have been evaluated individually by the grouper software. Following its clinical matching algorithms, two concurrently occurring episodes have been identified. The next step in the building of an episode requires the 'clustering' of expense lines with identical assignment. Clusters are the essential building blocks of an episode. This will be discussed in the next section.

56 Building the -Based Episode Building an episode requires more than clinical information alone. It requires an understanding of how, by whom, and in what venue health care services are provided. Towards that end, each expense line is evaluated, in addition to the clinical information, with respect to the type of health care provider responsible for the service. This information is crucial in developing clusters - the building blocks of an episode. Identifying Clusters: Anchor and Ancillary Records Other than the individual expense lines, the cluster is the smallest unit of analysis in the methodology. Each cluster is comprised of one anchor record and zero, one or more ancillary records. Further, each treatment episode consists of one or more clusters. Anchor records represent a service by a clinician engaging in the direct evaluation, management or treatment of a patient, such as office visits, surgery or certain therapies. The identification of an anchor record is a significant event because it represents that a clinician has evaluated a patient, and has decided on the types of services required to further identify and treat the patient's condition. Anchor records are identified by a combination of the billing provider and the procedure code billed. There are two types of anchor records. Management records for the most part consist of the evaluation and management CPT-4 codes; surgery records are the procedural CPT-4 codes. (These surgery records should not be confused with defining surgeries. Only a special subset of the surgery records are considered defining surgeries and can therefore shift an episode.) From a methodological perspective, only anchor records can identify the beginning of an episode, or for that matter, continue an episode with respect to the clean periods. Only a clinician can generate an anchor record. There is one notable exception to this rule: in certain instances a facility record can behave as an anchor record and begin an episode. Ancillary records represent services that are incidental to the direct evaluation, management and treatment of a patient. Examples of ancillary records are x-rays, pharmaceuticals and laboratory tests. Ancillary records are only identified by the procedure code billed. Therefore, ancillary records can be produced by both clinicians and non-clinicians. There are three types of ancillary records: facility records, which include the facility component of ambulatory surgery centers, and inpatient room and board services, pharmaceutical records, and ancillary records, which consist primarily of laboratory tests and x-rays. Each ancillary record links to one and only one anchor record, which by inference assigns the responsibility of that ancillary service to the clinician on the anchor record. The identification of anchor [management and surgery] and ancillary [facility, pharmaceutical and ancillary (x-ray and lab tests)] are added to the expense lines, designated by M, S, F, P, or A. This is shown in our example in Table 5 in the column labeled RECord Type. Note that our example has been slightly modified to display the provider type, record type and cluster number. In addition, the episode column has been modified to show episode number. The office visit (an evaluation and management CPT-4 code) has been designated "M", a type of anchor record, while the x-ray and laboratory services have been designated "A" for ancillary records. Table 5 EXPENSE LINES SHOWING RECORD TYPE AND CLUSTER Provider ICD-9 ICD-9 ICD-9 ICD-9 CPT-4 AMT DESC EPISODE REC CLUS MD MD MD 490 Bronchitis 490 Bronchitis 490 Bronchitis Diabetes Blank Blank $X Office Visit Diabetes Blank Blank $Y Chest x- ray Diabetes Blank Blank $Z Blood Glucose Bronchitis M Bronchitis A 001 A

57 The first two records share the same cluster designation, 001, and belong to the same episode, Subsequent records which belong to this episode will have sequential cluster numbers, i.e., 002, 003 etc., but still share the same episode number The third expense line with the Blood Glucose billed procedure will group to a different episode. Ungroupable Expense Lines Orphan Records Despite all efforts to assign every expense line to an -based episode, there will be instances in which grouping of an ancillary record is not possible. The first instance involves ancillary records which do not have associated anchor records. These "ungroupable" ancillary records are known as Orphan records. Orphan records for ancillary services such as lab and x-ray are assigned to 999, orphan ancillary record. Orphan pharmaceutical records are assigned to 991, orphan drug record. Mismatched Records A second instance of an ungroupable ancillary record is where the billed procedure is inconsistent with any of the diagnoses. An example would be a billing procedure for the setting of a broken bone with a single diagnosis of acute respiratory infection. Mis-matched records are assigned to 998, Inappropriate Dx-CPT4 matched record. Complete and Incomplete Episodes Identifying discretely occurring, clinically homogeneous episodes of care is one challenge that the methodology has successfully addressed. The identification for subsequent analysis of complete episodes is the other. The notion of a complete episode, while intuitively simple to envision, is a bit more complex in the reality of claims data. For example, available claims data for grouping will have a chronological beginning date, say incurred claims as of January 1, Does an episode for acute bronchitis with its first anchor record on January 3rd begin with this claim or have we 'joined the episode in progress'? In other words, the episode of acute bronchitis might have begun sometime earlier (prior to January 1, 1993) and the data to identify the exact beginning date are not available. Of course, the opposite is also true. With data available from January 1, 1993 through December 31, 1994, how can we know if a claim record incurred on December 21st for an existing episode of acute bronchitis is the end of the episode? The answer to both questions is that under certain circumstances we cannot know whether a claim record is actually the 'true' beginning or the 'true' end of an episode. A distinction must be made then, between episodes which are to be considered complete from those whose completeness cannot be determined. Clean and Unknown Episode Starts A clean start is defined as a situation where the "true" beginning date for an episode is known. The methodology identifies a clean start by comparing the incurred date of the first anchor record of an episode with beginning date of the database (or a member's beginning eligibility date, if known), with the episode's clean period. Consider the following example in Figure 2 for 381 (Acute Bronchitis, with comorbidity, Age less than 5) which has a clean period of 60 days. Figure 2 Clean and Unknown Episode Starts and Finishes

58 If the first anchor record for an episode in this is determined to be at date A which is less than 60 days beyond the database start date of January 1, 1993, we cannot discern whether we have a clean start or whether we have joined the episode in progress, hence an unknown start. However, if the service date of the first anchor record is anywhere beyond 60 days after the database start date (dates B, C or D), the episode is considered to have a clean start. Clean and Unknown Episode Finishes A clean finish is defined as situation where the true finish date for an episode is known. Analogous to the clean start which uses each 's clean period to determine a known beginning, a clean finish uses the same clean period to determine a known finish. Consider the previous example in Figure 2. If the service date of the last anchor record is at least 60 days before the database end date, (dates A, B, or C) then we consider the episode to have a clean finish. On the other hand, if the service date of the last anchor record is somewhere between the database end date and 60 days before it, (date D) then we determined the episode to be an unknown finish. The finish is considered unknown because we cannot say with certainty whether the episode is truly finished or whether additional services will be provided and the data has yet to be submitted. Full Year Episodes and Analysis It is fairly easy to identify and analyze acute condition episodes. Acute bronchitis, for example, can be diagnosed and treated within a fairly short period of time and subsequent analysis is straightforward. But how do we analyze more chronic illnesses when, by their very nature, a treatment episode can last a very long time, sometimes for the patient's entire lifetime? Here, the concept of clean and unknown episode starts and finishes is key to the understanding of what is referred to as full year episodes and how to analyze them. Consider the example in the figure below of an episode in 390 (Chronic bronchitis, with complication, with comorbidity) which has a clean period of 180 days. In this example, anchor records for this episode occur at dates A, B, C, D and E. Note that the treatment for this episode spans well over one year. Figure 3 Full Year Episodes Assume that the timeframe from each anchor record to the next is less than 180 days. So, The anchor record at date A is an unknown start; The anchor records at dates B and C (if either were the first anchor records in this episode) represent a clean start; The anchor records at dates D and E (if either were the last anchor records in this episode) represent an unknown finish. To restate the question: How do we analyze this episode which spans over a year and which has both an unknown start and an unknown finish? We can also expand on this question since it becomes clear that for many conditions, especially chronic ones, it may become difficult to rely solely on episodes with both clean starts and finishes for subsequent analysis. In fact, in the case of lifelong illnesses such as diabetes or hypertension, one could argue that both clean starts and finishes should rarely be observed in the data. The question then becomes: For what duration of time should we limit our analysis of those episodes which can essentially last 'forever'?

59 From the perspective of certain clinically defined analyses, one might not want to prescribe a limited length for an episode. Yet from the perspective of comparing one episode to another, such as for provider profiling for example, it is important to do so since exogenous factors often exist that may lead to erroneous conclusions. Consider the case of two cardiologists treating hypertensive patients: Dr. Smith with 80 percent of her patients 'in-plan' for two years and Dr. Jones with 80 percent of his patients 'in-plan' for only one year. From an episode perspective, Dr. Smith, on average, will probably have episodes of longer duration and greater expense than Dr. Jones, despite the fact that she may be treating her patients more efficiently. Having already adjusted for case mix by s, we further allow proper episode by episode comparison by limiting the length of each episode to one year. Such episodes which extend beyond one year and are subsequently limited to one year for analytical purposes are referred to as full year episodes. Using our previous example, consider the following types of episodes which could be used in a provider profile (or other) analysis: Complete episode with both a clean start and a clean finish, where the service date of the first anchor record is at date B and the last anchor record is at date C; A full year episode with a clean start but unknown finish. This situation occurs when the service date of the first anchor record is at date B and the last anchor record is at date E. The full year episode is then comprised of all claim information from date B to 365 days after date B; A full year episode with an unknown start but a clean finish. This situation occurs when the service date of the first anchor record is at date A and the last anchor record is found at date C. The full year episode is then comprised of all claim information from date A to 365 days after date A. A full year episode with both an unknown start and unknown finish. This situation occurs when the service date of the first anchor record is at date A and the last anchor record is at date E. The full year episode is then comprised of all claim information from date A to 365 days after date A. Outlier Status The determination of outlier status is another important feature of the methodology. An episode is considered a low outlier if it is below its -specific low outlier charge trim point. Similarly, there exists a high outlier cost trim point for each beyond which an episode is flagged as a high outlier. An episode could be an outlier due to one of several reasons: Inaccurate processing of, or coding on, the claim records. The coding problem could exist in the cost field, diagnoses, CPT codes, dates of service, etc. Inappropriate treatment which includes under- or overutilization. The episode is treated appropriately but is truly dissimilar from other episodes in the. Presumably if there were more variables available to define the s, such as lab test results, and there were enough such episodes, a separate would be defined. If the member's plan enrollment date is not available, the patient is presumed to be enrolled in the plan from the database start date. However, the patient may have enrolled in the plan after the database start date and a treatment episode may have started prior to the patient's enrollment. Thus, the episode may become a low outlier due to incomplete claims availability when it really should be an episode with an unknown start date. NON-EPISODIC CATEGORIES The smallest unit of an episode of care is a cluster, and a cluster requires an anchor record. However there are many cases where drug and ancillary data are present for a member but no physician claims exist to create the anchors. Sometimes a matching physician claim doesn t exist because of a call-in prescription where the provider chooses not to submit a claim. Sometimes the number of refills of a prescription are so large that a given prescription claim may fall well beyond a reasonable period to match it to the ordering physician in the claims data. In addition, managed care is now replete with prevention activities which do not fall neatly into the concept of an episode of care. Such services however are extremely

60 important for the overall health of the member and cost containment for the health plan. The methodology is able to classify most of these data into clinically homogenous, non-episodic s. One set of non-episodic s are the screenings and immunizations which are incidental to other services. There are 9 such s, and they exist in their own major practice category. They range from glaucoma screenings to cholesterol tests. The other non-episodic category is the many groupings of pharmacy data with no provider intervention data. For those drugs (such as insulin) were the clinical category is obvious, they are bucketed into clinically homogenous groups. Most major practice categories have at least one such grouping of pharmacy data. Endocrinology has the most unique categories at 10. These non-episodic s are not assigned a unique episode number. The user can analyze these data separately from the episodic data. They serve both to illustrate the way medicine is practiced today as well as provide an opportunity to question if more physician monitoring and intervention may be appropriate. EXAMPLE APPLICATION: PROFILING REPORT CARD Provider ID: Michael Ashton, M.D West University Blvd., Suite 8b Ann Arbor, MN Number Average Episodes Cost PROVIDER PROFILING Casemix Index: 0.82 Performance Index: 1.17 Expected Cost Difference Total Difference English Description (3) (840) Routine Exam (8) 1,960 Other ENT Infection, w/o surgery ,855 Otitis media, with minor surgery (17) (1,394) Acute bronchitis, w/o comorbidity, age (11) (594) Minor inflammation of skin & subcutaneous tissue Asthma w/o comorbidity, age < (12) (600) Otitis media, w/o surgery Total 1, $27,450 This provider "report card" displays an overview of a Dr. Ashton's case mix adjusted performance, and in addition, offers specific feedback regarding his actual case mix. The Case mix Index is a measure of the morbidity underlying his case mix while the Performance Index measures overall financial performance (after adjusting for case mix). The higher the Case mix Index, the greater the patient severity; the higher the Performance Index, the greater the cost inefficiency. In this example, after adjusting for his relatively easy case mix (0.82), Dr. Ashton costs 17% more to treat his patients than other providers treating an identical case mix. This relative inefficiency is due primarily to one : Otitis Media, with minor surgery. Subsequent drill-down analysis can uncover the source of this cost inefficiency.

61 EXAMPLE APPLICATION: PHARMACEUTICAL-BASED DISEASE MANAGEMENT Analyzing an -defined episode for component prescription drug costs can be illuminating. A good example is the study of prescribing patterns in the treatment of depression episodes ( 88, sub 2; Mood disorder, depressed, without psychosis). An analysis of completed episodes using the Symmetry multi-tiered drug classification system yields the expected results. Virtually all prescriptions within these episodes (99.85%) were for Central Nervous System Agents (classification at the TCC level). About half of those (49.17%) were SSRI Antidepressants (classification at the PCC level). Figure 4 illustrates the breakdown of these SSRI medications (classification at the DCC level). Figure 4 Distribution of SSRI Drug Costs for Depression Episodes The cost effectiveness of pharmaceutical interventions can be easily determined by stratifying episode cost by drug. In the example below, note that the pharmaceutical cost for Drug Z is the most costly but results in the most inexpensive episode. Why? Because other costs associated with the episode are reduced. What does this mean? Perhaps Drug Z should make its way to your plan's drug formulary. Result: work smarter instead of harder. Figure 5 Relative Difference in Episode Cost, by Type, for 329: Otitis Media, w/o Surgery

The BlueCompare SM Physician Designation Program

The BlueCompare SM Physician Designation Program The BlueCompare SM Physician Designation Program Continuing escalation of health care costs has driven premiums and medical expenses to higher and higher levels each year, motivating employers and consumers

More information

Cigna Care Designation and Physician Quality and Cost-Efficiency Displays 2016 Methodologies Whitepaper

Cigna Care Designation and Physician Quality and Cost-Efficiency Displays 2016 Methodologies Whitepaper Cigna Care Designation and Physician Quality and Cost-Efficiency Displays 2016 Methodologies Whitepaper For Health Care Professionals February 2016 Introduction... 2 Cigna Physician Quality and Cost-Efficiency

More information

Clinical Quality Measures Summary of Upcoming Enhancements

Clinical Quality Measures Summary of Upcoming Enhancements Upcoming coding enhancements will impact the logic behind the clinical quality indicators applicable to your practice specialty. Please refer to this grid for a summary of the coding enhancements and some

More information

Florida Blue QUALITY PERFORMANCE METRIC STANDARDS FEBRUARY 2013

Florida Blue QUALITY PERFORMANCE METRIC STANDARDS FEBRUARY 2013 Florida Blue QUALITY PERFORMANCE METRIC STANDARDS FEBRUARY 2013 QUALITY PERFORMANCE METRIC CALCULATION QUALITY METRICS SELECTED FOR MEASUREMENT Per Section 3.2 of the Agreement, HCPP must meet the following

More information

UnitedHealth Premium Program Case-mix Adjusted Benchmarks with Severity or Risk Adjustment

UnitedHealth Premium Program Case-mix Adjusted Benchmarks with Severity or Risk Adjustment UnitedHealth Premium Program Case-mix Adjusted Benchmarks with Severity or Risk Adjustment Resources u Phone: 866-270-5588 u u Website: UHCprovider.com/Premium Mail: UnitedHealthcare - UnitedHealth Premium

More information

Arkansas Health Care Payment Improvement Initiative Congestive Heart Failure Algorithm Summary

Arkansas Health Care Payment Improvement Initiative Congestive Heart Failure Algorithm Summary Arkansas Health Care Payment Improvement Initiative Congestive Heart Failure Algorithm Summary Congestive Heart Failure Algorithm Summary v1.2 (1/5) Triggers PAP assignment Exclusions Episode time window

More information

CLINICAL PROCESS IMPROVEMENT INITIATIVE (CPII) EFFICIENCY REPORT EXPLANATION January 4, 2016

CLINICAL PROCESS IMPROVEMENT INITIATIVE (CPII) EFFICIENCY REPORT EXPLANATION January 4, 2016 CLINICAL PROCESS IMPROVEMENT INITIATIVE (CPII) EFFICIENCY REPORT EXPLANATION January 4, 2016 WHAT IS AN EPISODE OF CARE? An episode of care is a grouping of a patient s health care claims for a unique

More information

UnitedHealth Premium Designation Program Case-mix Adjusted Benchmarks with Severity or Risk Adjustment

UnitedHealth Premium Designation Program Case-mix Adjusted Benchmarks with Severity or Risk Adjustment UnitedHealth Premium Designation Program Case-mix Adjusted Benchmarks with Severity or Risk Adjustment Resources u Phone: 866-270-5588 u u u Website: UnitedHealthcareOnline.com > Quick Links > UnitedHealth

More information

HEDIS QUICK REFERENCE GUIDE: DOCUMENTATION TIPS FOR ADULT MEASURES

HEDIS QUICK REFERENCE GUIDE: DOCUMENTATION TIPS FOR ADULT MEASURES HEDIS QUICK REFERENCE GUIDE: DOCUMENTATION TIPS FOR ADULT MEASURES For Health Care Providers January 2018 Helping you improve your scores, as you improve the health of your patients. Healthcare Effectiveness

More information

Technical Appendix for Outcome Measures

Technical Appendix for Outcome Measures Study Overview Technical Appendix for Outcome Measures This is a report on data used, and analyses done, by MPA Healthcare Solutions (MPA, formerly Michael Pine and Associates) for Consumers CHECKBOOK/Center

More information

2018 National Oncologists Workforce Study OCTOBER 2018

2018 National Oncologists Workforce Study OCTOBER 2018 2018 National Oncologists Workforce Study OCTOBER 2018 Introduction Providers have long cautioned that the U.S. healthcare system is facing imminent physician shortages across many different specialties.

More information

Arkansas Blue Cross and Blue Shield (ABCBS) Patient Centered Medical Home (PCMH) Specifications Manual

Arkansas Blue Cross and Blue Shield (ABCBS) Patient Centered Medical Home (PCMH) Specifications Manual Arkansas Blue Cross and Blue Shield (ABCBS) Patient Centered Medical Home (PCMH) Specifications Manual 2017 This document is a guide to the 2017 Arkansas Blue Cross and Blue Shield Patient-Centered Medical

More information

UnitedHealth Premium Program Case-mix, Severity and Risk

UnitedHealth Premium Program Case-mix, Severity and Risk UnitedHealth Premium Program Case-mix, Severity and Risk Resources u Phone: 866-270-5588 u u Website: UnitedHealthPremium.UHC.com Mail: UnitedHealthcare - UnitedHealth Premium Program MN017-W700 9700 Health

More information

Efficiency Methodology

Efficiency Methodology Efficiency Methodology David C. Schutt, MD Bob Kelley Thomson Healthcare October 2007 Overview Definition Clinical Grouping Methods Implementation Considerations Reporting to Physician Organizations Example

More information

Quality Corp Measures Description and Methodologies

Quality Corp Measures Description and Methodologies Quality Corp Measures Description and Methodologies Overview: The Oregon Health Care Quality Corporation (Q Corp) is dedicated to improving the quality and affordability of health care in Oregon by leading

More information

Key Quality of Care Measures. Blue Cross Blue Shield of Michigan Traditional, PPO and POS Members. Fourth Quarter 2003

Key Quality of Care Measures. Blue Cross Blue Shield of Michigan Traditional, PPO and POS Members. Fourth Quarter 2003 Key Quality of Care Measures Blue Cross Blue Shield of Michigan Traditional, PPO and POS Members Fourth Quarter 2003 Blue Cross Blue Shield of Michigan is a nonprofit corporation and independent licensee

More information

Family Planning Eligibility Program

Family Planning Eligibility Program INDIANA HEALTH COVERAGE PROGRAMS PROVIDER REFERENCE M ODULE Family Planning Eligibility Program L I B R A R Y R E F E R E N C E N U M B E R : P R O M O D 0 0 0 5 3 P U B L I S H E D : N O V E M B E R 2

More information

Care1st Health Plan Taking Quality to the Next Level REPORTING YEAR HEDIS Summary - MPL (Measurement Year 2012)

Care1st Health Plan Taking Quality to the Next Level REPORTING YEAR HEDIS Summary - MPL (Measurement Year 2012) Care1st Health Plan s Quality Improvement Department has been diligently working towards improving the Healthcare Effectiveness Data and Information Sets (HEDIS) results across all lines of business. HEDIS

More information

Management of Heart Failure: Review of the Performance Measures by the Performance Measurement Committee of the American College of Physicians

Management of Heart Failure: Review of the Performance Measures by the Performance Measurement Committee of the American College of Physicians Performance Measurement Management of Heart Failure: Review of the Performance Measures by the Performance Measurement Committee of the American College of Physicians Writing Committee Amir Qaseem, MD,

More information

TECHNICAL NOTES APPENDIX SUMMER

TECHNICAL NOTES APPENDIX SUMMER TECHNICAL NOTES APPENDIX SUMMER Hospital Performance Report Summer Update INCLUDES PENNSYLVANIA INPATIENT HOSPITAL DISCHARGES FROM JULY 1, 2006 THROUGH JUNE 30, 2007 The Pennsylvania Health Care Cost Containment

More information

THE NATIONAL QUALITY FORUM

THE NATIONAL QUALITY FORUM THE NATIONAL QUALITY FORUM National Voluntary Consensus Standards for Patient Outcomes Table of Measures Submitted-Phase 1 As of March 5, 2010 Note: This information is for personal and noncommercial use

More information

TOTAL HIP AND KNEE REPLACEMENTS. FISCAL YEAR 2002 DATA July 1, 2001 through June 30, 2002 TECHNICAL NOTES

TOTAL HIP AND KNEE REPLACEMENTS. FISCAL YEAR 2002 DATA July 1, 2001 through June 30, 2002 TECHNICAL NOTES TOTAL HIP AND KNEE REPLACEMENTS FISCAL YEAR 2002 DATA July 1, 2001 through June 30, 2002 TECHNICAL NOTES The Pennsylvania Health Care Cost Containment Council April 2005 Preface This document serves as

More information

Quality Corp Measures Description and Methodologies

Quality Corp Measures Description and Methodologies Quality Corp Measures Description and Methodologies Overview: The Oregon Health Care Quality Corporation (Q Corp) is dedicated to improving the quality and affordability of health care in Oregon by leading

More information

Supplementary Online Content

Supplementary Online Content Supplementary Online Content Friedberg MW, Rosenthal MB, Werner RM, Volpp KG, Schneider EC. Effects of a medical home and shared savings intervention on quality and utilization of care. Published online

More information

Professional Non Covered Codes Policy

Professional Non Covered Codes Policy Policy Number 2018R7102I Professional Non Covered Codes Policy Annual Approval Date 7/12/2017 Approved By Reimbursement Policy Oversight Committee IMPORTANT NOTE ABOUT THIS REIMBURSEMENT POLICY You are

More information

QIP/HEDIS Measure Webinar Series

QIP/HEDIS Measure Webinar Series QIP/HEDIS Measure Webinar Series September 26, 2017 Presenters: Partnership HealthPlan Quality Department Partnership HealthPlan of California To avoid echoes and feedback, we request that you use the

More information

HEDIS. Quick Reference Guide. For more information, visit

HEDIS. Quick Reference Guide. For more information, visit HEDIS Quick Reference Guide For more information, visit www.ncqa.org HEDIS Quick Reference Guide Updated to reflect NCQA HEDIS 2017 Technical Specifications Michigan Complete Health Medicare-Medicaid Plan

More information

2015 Reimbursement Information for Mammography, CAD and Digital Breast Tomosynthesis 1

2015 Reimbursement Information for Mammography, CAD and Digital Breast Tomosynthesis 1 GE Healthcare 2015 Reimbursement Information for Mammography, CAD and Digital Breast Tomosynthesis 1 April, 2015 www.gehealthcare.com/reimbursement This advisory addresses Medicare coding, coverage and

More information

SUMMARY TABLE OF MEASURES, PRODUCT LINES AND CHANGES

SUMMARY TABLE OF MEASURES, PRODUCT LINES AND CHANGES Summary Table of Measures, Product Lines and Changes 1 SUMMARY TABLE OF MEASURES, PRODUCT LINES AND CHANGES General Guidelines for Data Collection and Reporting Guidelines for Calculations and Sampling

More information

Trends in Health Plan Design: High Performance Networks. John Bertko, F.S.A., MAAA Vice President and Chief Actuary Humana Inc July 28, 2005

Trends in Health Plan Design: High Performance Networks. John Bertko, F.S.A., MAAA Vice President and Chief Actuary Humana Inc July 28, 2005 Trends in Health Plan Design: High Performance Networks John Bertko, F.S.A., MAAA Vice President and Chief Actuary Humana Inc July 28, 2005 1 Overview What is a High Performance Network (HPN) Tighter network

More information

Reject Code Reason for Rejection What to do

Reject Code Reason for Rejection What to do Reject Code Reason for Rejection What to do 10 Hospital where services rendered missing or invalid. Input the Hospital where services were rendered on the HCFA. 11 Patient first name missing or invalid.

More information

Performance Analysis:

Performance Analysis: Performance Analysis: Healthcare Utilization of CCNC- Population 2007-2010 Prepared by Treo Solutions JUNE 2012 Table of Contents SECTION ONE: EXECUTIVE SUMMARY 4-5 SECTION TWO: REPORT DETAILS 6 Inpatient

More information

Quality Care Plus 2015 Primary Care Physician Incentive Program. Now includes Medicare patients!

Quality Care Plus 2015 Primary Care Physician Incentive Program. Now includes Medicare patients! Quality Care Plus 2015 Primary Care Physician Incentive Program Now includes Medicare patients! Health Partners Plans (HPP) would like to express our appreciation for the invaluable role our primary care

More information

preventive health care measure

preventive health care measure introduction 98 < pre- & post-esrd preventive care 1 < diabetic care 12 < cancer screening 14 < chapter summary 16 Chapter Five preventive health care measure SSo neither ought you to attempt to cure the

More information

Volunteering in Oklahoma City, OK

Volunteering in Oklahoma City, OK 6/17/2010 Oklahoma City Profile - Volunteering in information on volunteering and civic engagement Volunteering in Oklahoma City, OK Statistics for this area were collected within the Oklahoma City Metropolitan

More information

Quality ID #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care

Quality ID #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care Quality ID #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care 2018 OPTIONS FOR INDIVIDUAL MEASURES: REGISTRY ONLY MEASURE TYPE: Process DESCRIPTION: Percentage

More information

The Cost Burden of Worsening Heart Failure in the Medicare Fee For Service Population: An Actuarial Analysis

The Cost Burden of Worsening Heart Failure in the Medicare Fee For Service Population: An Actuarial Analysis Client Report Milliman Client Report The Cost Burden of Worsening Heart Failure in the Medicare Fee For Service Population: An Actuarial Analysis Prepared by Kathryn Fitch, RN, MEd Principal and Healthcare

More information

HEDIS Quick Reference Guide Updated to reflect NCQA HEDIS 2016 Technical Specifications

HEDIS Quick Reference Guide Updated to reflect NCQA HEDIS 2016 Technical Specifications HEDIS Quick Reference Guide Updated to reflect NCQA HEDIS 2016 Technical Specifications Fidelis SecureCare strives to provide quality healthcare to our membership as measured through HEDIS quality metrics.

More information

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 4 Episodes

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 4 Episodes TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 4 Episodes Attention deficit hyperactivity disorder (ADHD); Opposition defiance disorder (ODD); Coronary artery bypass

More information

Partial Hospitalization Program Program for Evaluating Payment Patterns Electronic Report. User s Guide Sixth Edition. Prepared by

Partial Hospitalization Program Program for Evaluating Payment Patterns Electronic Report. User s Guide Sixth Edition. Prepared by Partial Hospitalization Program Program for Evaluating Payment Patterns Electronic Report User s Guide Sixth Edition Prepared by Partial Hospitalization Program Program for Evaluating Payment Patterns

More information

Montgomery Cares Clinical Performance Measures

Montgomery Cares Clinical Performance Measures Montgomery Cares Clinical Performance Measures Fiscal Year 2012 December 13, 2012 1 Montgomery Cares Clinical Performance Measures, FY 2012 Table of Contents Executive Summary 2 Background. 3 Results Reporting

More information

Impact of WRVU Changes. Allowed Charges (Millions)

Impact of WRVU Changes. Allowed Charges (Millions) Key Financial and Operational s from the Proposed 2018 PFS Rule: The 2018 Physician Fee Schedule (PFS) proposed rule was made available on July 13, 2018. A detailed summary of the rule will be available

More information

Precyse University ICD-10 Education Tracks

Precyse University ICD-10 Education Tracks Precyse University ICD-10 Education Tracks The following information will help you understand the Precyse University ICD-10 Educational Tracks. Each impacted population requires specific education to prepare

More information

A Prospective Comparison of Two Commercially Available Hospital Admission Risk Scores

A Prospective Comparison of Two Commercially Available Hospital Admission Risk Scores A Prospective Comparison of Two Commercially Available Hospital Admission Risk Scores Dale Eric Green, MD MHA FCCP: Chief Medical Information Officer, Cornerstone Health Care AMGA April, 2014 Disclosure

More information

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 2 Episodes

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 2 Episodes TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 2 Episodes Acute COPD exacerbation (COPD); Screening and surveillance colonoscopy (COL); and Outpatient and non-acute

More information

Combining Risk Adjustment and HEDIS to Improve Quality of Care. Colleen Gianatasio, CPC, CPC-P, CPMA, CPC-I, CRC

Combining Risk Adjustment and HEDIS to Improve Quality of Care. Colleen Gianatasio, CPC, CPC-P, CPMA, CPC-I, CRC Combining Risk Adjustment and HEDIS to Improve Quality of Care Colleen Gianatasio, CPC, CPC-P, CPMA, CPC-I, CRC Agenda Improving primary care in today s health care environment Risk adjustment basics (using

More information

DESCRIPTION: Percentage of women years of age who had a mammogram to screen for breast cancer

DESCRIPTION: Percentage of women years of age who had a mammogram to screen for breast cancer Quality ID #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care 2018 OPTIONS FOR INDIVIDUAL MEASURES: CLAIMS ONLY MEASURE TYPE: Process DESCRIPTION: Percentage

More information

Anthem Pay-for- Performance (HEDIS )*

Anthem Pay-for- Performance (HEDIS )* Serving Hoosier Healthwise, Healthy Indiana Plan Anthem Pay-for- Performance (HEDIS )* [Candace Adye, RN Amanda Gonzalez, RN] *HEDIS is a registered trademark of the National Committee for Quality Assurance

More information

ACO Congress Conference Pre Session Clinical Performance Measurement

ACO Congress Conference Pre Session Clinical Performance Measurement ACO Congress Conference Pre Session Clinical Performance Measurement Lynne Rothney-Kozlak, MPH Interim VP, ACO Collaborative (Independent Consultant) October 25, 2010 Agenda for Presentation 1. The Framework

More information

TECHNICAL NOTES APPENDIX SUMMER

TECHNICAL NOTES APPENDIX SUMMER TECHNICAL NOTES APPENDIX SUMMER Hospital Performance Report Summer Update INCLUDES PENNSYLVANIA INPATIENT HOSPITAL DISCHARGES FROM July 1, 2005 through June 30, 2006 The Pennsylvania Health Care Cost Containment

More information

2018 P4P Overview 0518.PR.P.PP.1 6/18

2018 P4P Overview 0518.PR.P.PP.1 6/18 2018 P4P Overview Agenda MHS Pay For Performance (P4P) Ambetter P4P Program Secure Web Reporting Question and Answer What You Will Learn 1. Measure Overviews & Specifications 2. Documentation Requirements

More information

NATIONAL HEALTH POLICY FORUM

NATIONAL HEALTH POLICY FORUM NATIONAL HEALTH POLICY FORUM UTILIZATION PATTERNS IN ADVANCED IMAGING: THE RADIOLOGY BENEFIT MANAGEMENT EXPERIENCE Susan Nedza, MD, MBA, FACEP Chief Medical Officer AIM Specialty Health / WellPoint, Inc.

More information

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 5 Episodes

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 5 Episodes TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 5 Episodes Mastectomy, Breast Cancer Medical Oncology, Breast Biopsy, Tonsillectomy, Otitis media, Anxiety, Non-emergent

More information

Value of Hospice Benefit to Medicaid Programs

Value of Hospice Benefit to Medicaid Programs One Pennsylvania Plaza, 38 th Floor New York, NY 10119 Tel 212-279-7166 Fax 212-629-5657 www.milliman.com Value of Hospice Benefit May 2, 2003 Milliman USA, Inc. New York, NY Kate Fitch, RN, MEd, MA Bruce

More information

The Role of HMOs in Managing Diabetes Technical Notes

The Role of HMOs in Managing Diabetes Technical Notes The Role of HMOs in Managing Diabetes Technical Notes The 225 Market Street, Suite 400 Harrisburg, PA 17101 (717) 232-6787 www.phc4.org Foreword These Technical Notes are intended to accompany the public

More information

Advances in Alignment, Measurement, and Performance MY 2017 Results Highlights

Advances in Alignment, Measurement, and Performance MY 2017 Results Highlights Advances in Alignment, Measurement, and Performance MY 2017 Results Highlights Align. Measure. Perform. (AMP) Programs Launched in 2003, VBP4P is a statewide performance improvement program and one of

More information

Care Facilitation Quality Improvement Report

Care Facilitation Quality Improvement Report Disease Management Program Clinical Outcomes for Reporting Period: 2006 Diabetes 100.0% 90.0% % of participants with diabetes 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% % participants with at

More information

Merit-based Incentive Payment System (MIPS): Cost Measure Field Test Reports Fact Sheet

Merit-based Incentive Payment System (MIPS): Cost Measure Field Test Reports Fact Sheet Merit-based Incentive Payment System (MIPS): Cost Measure Field Test Reports Fact Sheet The Quality Payment Program The Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) established the Quality

More information

AMCP Webinar Series. Exchanges and Qualified Health Plans: How your voice can shape the future of quality reporting 14 January 2014.

AMCP Webinar Series. Exchanges and Qualified Health Plans: How your voice can shape the future of quality reporting 14 January 2014. AMCP Webinar Series Exchanges and Qualified Health Plans: How your voice can shape the future of quality reporting 14 January 2014 Speaker Mitzi Wasik, Pharm.D., BCPS Director, Pharmacy Medicare Programs

More information

Troubleshooting Audio

Troubleshooting Audio Welcome Audio for this event is available via ReadyTalk Internet streaming. No telephone line is required. Computer speakers or headphones are necessary to listen to streaming audio. Limited dial-in lines

More information

November Data Jam. Mastering PSYCKES: Maximizing Multiple Data Sources to Operationalize a Population Health Approach

November Data Jam. Mastering PSYCKES: Maximizing Multiple Data Sources to Operationalize a Population Health Approach November Data Jam Mastering PSYCKES: Maximizing Multiple Data Sources to Operationalize a Population Health Approach Anni Kramer, LMSW & Erica Van De Wal-Ward, MA New York Office of Mental Health Anni

More information

APR DRG Data Discovery

APR DRG Data Discovery APR DRG Data Discovery Henry Johnson MD, MPH Midas+ Vice President and Medical Director Vanessa Dorr RN MSN Midas+ DataVision Clinical Consultant Review: Agenda 3M APR DRG Methodology DataVision: Web and

More information

Kathryn Goodwin, Senior Project Manager, and Karen Johnson, Senior Director

Kathryn Goodwin, Senior Project Manager, and Karen Johnson, Senior Director TO: FR: RE: Consensus Standards Approval Committee (CSAC) Kathryn Goodwin, Senior Project Manager, and Karen Johnson, Senior Director Musculoskeletal Off-Cycle Measure Review DA: April 26, 2017 The CSAC

More information

STATE OPERATIONS MANUAL

STATE OPERATIONS MANUAL STATE OPERATIONS MANUAL Appendix W Survey Protocol, Regulations and Interpretive Guidelines for Critical Access Hospitals (CAHs) And Swing-Beds in CAHs Revisions 84, 06-07-21013 C-0151 Physician Ownership

More information

Modular Program Report

Modular Program Report Modular Program Report Disclaimer The following report(s) provides findings from an FDA initiated query using Sentinel. While Sentinel queries may be undertaken to assess potential medical product safety

More information

Benefits and Harms of Routine Preoperative Testing: A Comparative Effectiveness Review

Benefits and Harms of Routine Preoperative Testing: A Comparative Effectiveness Review Benefits and Harms of Routine Preoperative Testing: A Comparative Effectiveness Review Brown Evidence- based Practice Center, Brown University School of Public Health Ethan M. Balk, MD, MPH Amy Earley,

More information

Performance Measurement: HEDIS, STARS and More. Margaret E. O Kane NCQA President CAPG Educational Series October 27, 2016

Performance Measurement: HEDIS, STARS and More. Margaret E. O Kane NCQA President CAPG Educational Series October 27, 2016 Performance Measurement: HEDIS, STARS and More Margaret E. O Kane NCQA President CAPG Educational Series October 27, 2016 Defining Quality & Performance Trends The Triple Aim and National Priorities Partnership

More information

2019 CMS Web Interface

2019 CMS Web Interface CMS Web Interface PREV-5 (NQF 2372): Breast Cancer Screening Measure Steward: NCQA CMS Web Interface V3.0 Page 1 of 18 xx/xx/2018 Contents INTRODUCTION... 3 CMS WEB INTERFACE SAMPLING INFORMATION... 4

More information

Measure Owner Designation. AMA-PCPI is the measure owner. NCQA is the measure owner. QIP/CMS is the measure owner. AMA-NCQA is the measure owner

Measure Owner Designation. AMA-PCPI is the measure owner. NCQA is the measure owner. QIP/CMS is the measure owner. AMA-NCQA is the measure owner 2011 EHR Measure Specifications The specifications listed in this document have been updated to reflect clinical practice guidelines and applicable health informatics standards that are the most current

More information

QBPC Claims Based Provider Quick Reference Guide

QBPC Claims Based Provider Quick Reference Guide QBPC Claims Based Provider Quick Reference Guide Category: Diabetes Chronic Suite ICD-10-CM diagnosis HbA1c Test Codes LOINC Evidence of Treatment for Nephropathy Codes E10; E11; E13 83036-37 17856-6,

More information

Symmetry Episode Treatment Groups

Symmetry Episode Treatment Groups Symmetry Episode Treatment Groups Measuring health care with meaningful episodes of care Identifying clinical episodes of illness and the services involved in their diagnosis, management, and treatment

More information

WCHQ MEASURES AT A GLANCE

WCHQ MEASURES AT A GLANCE WCHQ Ambulatory Measures A1C Blood Sugar Testing A1C Blood Sugar Control Patients with diabetes Patients with diabetes office visit in. Gestational Diabetes (code 648.8) is office visit in. Compliance

More information

Section II: Detailed Measure Specifications

Section II: Detailed Measure Specifications Section II: Detailed Measure Specifications Provide sufficient detail to describe how a measure would be calculated from the recommended data sources, uploading a separate document (+ Upload attachment)

More information

Measure #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care

Measure #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care Measure #112 (NQF 2372): Breast Cancer Screening National Quality Strategy Domain: Effective Clinical Care 2017 OPTIONS FOR INDIVIDUAL MEASURES: CLAIMS ONLY MEASURE TYPE: Process DESCRIPTION: Percentage

More information

Anthem Blue Precision Quality Designation Methodology Summary 2017

Anthem Blue Precision Quality Designation Methodology Summary 2017 This document outlines Anthem s Blue Precision methodology for measuring Physician Quality. Anthem has physician quality transparency programs for physicians practicing in the following specialties: Allergy/Immunology,

More information

QUALITY IMPROVEMENT Section 9

QUALITY IMPROVEMENT Section 9 Quality Improvement Program The Plan s Quality Improvement Program serves to improve the health of its members through emphasis on health maintenance, education, diagnostic testing and treatment. The Quality

More information

2018 CMS Web Interface

2018 CMS Web Interface CMS Web Interface PREV-5 (NQF 2372): Breast Cancer Screening Measure Steward: NCQA CMS Web Interface V2.1 Page 1 of 18 06/25/ Contents INTRODUCTION... 3 CMS WEB INTERFACE SAMPLING INFORMATION... 4 BENEFICIARY

More information

CASE-MIX ANALYSIS ACROSS PATIENT POPULATIONS AND BOUNDARIES: A REFINED CLASSIFICATION SYSTEM DESIGNED SPECIFICALLY FOR INTERNATIONAL USE

CASE-MIX ANALYSIS ACROSS PATIENT POPULATIONS AND BOUNDARIES: A REFINED CLASSIFICATION SYSTEM DESIGNED SPECIFICALLY FOR INTERNATIONAL USE CASE-MIX ANALYSIS ACROSS PATIENT POPULATIONS AND BOUNDARIES: A REFINED CLASSIFICATION SYSTEM DESIGNED SPECIFICALLY FOR INTERNATIONAL USE A WHITE PAPER BY: ROBERT MULLIN, MD JAMES VERTREES, PHD RICHARD

More information

Anesthesia Reimbursement

Anesthesia Reimbursement This drafted policy is open for a two-week public comment period. This box is not part of the drafted policy language itself, and is intended for use only during the comment period as a means to provide

More information

HEALTHCARE REFORM. September 2012

HEALTHCARE REFORM. September 2012 HEALTHCARE REFORM Accountable Care Organizations: ACOs 101 September 2012 The enclosed slides are intended to provide you with a general overview of accountable care organizations (ACOs), created within

More information

Performance Improvement Projects Related to CDC s 6 18 Initiative: A Scan of External Quality Review Organization Reports

Performance Improvement Projects Related to CDC s 6 18 Initiative: A Scan of External Quality Review Organization Reports Technical Assistance Tool June 2018 Performance Improvement Projects Related to CDC s 6 18 Initiative: A Scan of External Quality Review Organization Reports S tates implementing interventions under CDC

More information

Chapter 15 Section 1

Chapter 15 Section 1 Chapter 15 Section 1 Issue Date: November 6, 2007 Authority: 32 CFR 199.14(a)(3) and (a)(6)(ii) 1.0 APPLICABILITY This policy is mandatory for the reimbursement of services provided either by network or

More information

2018 Cerebrovascular Reimbursement Coding Fact Sheet

2018 Cerebrovascular Reimbursement Coding Fact Sheet The information contained in this document is provided for informational purposes only and represents no statement, promise, or guarantee by Cordis Corporation concerning levels of reimbursement, payment,

More information

Appendix C NEWBORN HEARING SCREENING PROJECT

Appendix C NEWBORN HEARING SCREENING PROJECT Appendix C NEWBORN HEARING SCREENING PROJECT I. WEST VIRGINIA STATE LAW All newborns born in the State of West Virginia must be screened for hearing impairment as required in WV Code 16-22A and 16-1-7,

More information

Arkansas Health Care Payment Improvement Initiative Percutaneous Coronary Intervention Algorithm Summary

Arkansas Health Care Payment Improvement Initiative Percutaneous Coronary Intervention Algorithm Summary Arkansas Health Care Payment Improvement Initiative Percutaneous Coronary Intervention Algorithm Summary Percutaneous Coronary Intervention (PCI) Algorithm Summary v1.0 Page 2 of 8 Triggers PAP assignment

More information

Hospital Discharge Data

Hospital Discharge Data Hospital Discharge Data West Virginia Health Care Authority Hospitalization data were obtained from the West Virginia Health Care Authority s (WVHCA) hospital discharge database. Data are submitted by

More information

Date: April 26, Proposed changes are highlighted in yellow in the attached table.

Date: April 26, Proposed changes are highlighted in yellow in the attached table. Date: April 26, 2013 Re: 2014 Statewide Quality Reporting and Measurement System: MN Community Measurement s Preliminary Recommendations for Physician Clinic and Ambulatory Surgical Center Measures Minnesota

More information

UNDERSTANDING NEW HAMPSHIRE DENTAL CLAIMS DATA

UNDERSTANDING NEW HAMPSHIRE DENTAL CLAIMS DATA UNDERSTANDING NEW HAMPSHIRE DENTAL CLAIMS DATA Quality Assurance and Preliminary Analysis Institute for Health Policy and Practice at UNH Abbott Willard, MS MPH Jo Porter, MPH Amy Costello, MPH Funding

More information

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 3 Episodes

TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 3 Episodes TENNCARE Bundled Payment Initiative: Description of Bundle Risk Adjustment for Wave 3 Episodes Respiratory Infection (RI); Pneumonia (PNA); Inpatient Urinary Tract Infection (UTI-I; Outpatient Urinary

More information

Integrated Health Care Economics

Integrated Health Care Economics Original Research Integrated Health Care Economics Part 1: Are Specialty Physician Services Revenues Reliable Predictors of Community Health System Financial Performance? By Daniel K. Zismer, PhD, Jeffrey

More information

ACO #44 Use of Imaging Studies for Low Back Pain

ACO #44 Use of Imaging Studies for Low Back Pain Measure Information Form (MIF) DATA SOURCE Medicare Claims Medicare beneficiary enrollment data MEASURE SET ID ACO #44 VERSION NUMBER AND EFFECTIVE DATE Version 1, effective 01/01/18 CMS APPROVAL DATE

More information

Get the Right Reimbursement for High Risk Patients

Get the Right Reimbursement for High Risk Patients Get the Right Reimbursement for High Risk Patients A Proven Strategy for Managing Hierarchical Condition Categories (HCC) in your EHR 847-272-1242 sales@e-imo.com e-imo.com 1 OVERVIEW Medicare Advantage

More information

Accelero Identifies Opportunities to Provide Greater Value in Hip Fracture Care

Accelero Identifies Opportunities to Provide Greater Value in Hip Fracture Care Accelero Health Partners, 2015 WHITE PAPER Accelero Identifies Opportunities to Provide Greater Value in Hip Fracture Care Jason Pry, Senior Director ABSTRACT Every year more than a quarter of a million

More information

Clinical Policy Title: Fluorescence in situ hybridization for cervical cancer screening

Clinical Policy Title: Fluorescence in situ hybridization for cervical cancer screening Clinical Policy Title: Fluorescence in situ hybridization for cervical cancer screening Clinical Policy Number: 01.01.02 Effective Date: April 1, 2015 Initial Review Date: January 21, 2015 Most Recent

More information

2018 OPTIONS FOR INDIVIDUAL MEASURES: REGISTRY ONLY. MEASURE TYPE: Process

2018 OPTIONS FOR INDIVIDUAL MEASURES: REGISTRY ONLY. MEASURE TYPE: Process Quality ID #134 (NQF 0418): Preventive Care and Screening: Screening for Depression and Follow-Up Plan National Quality Strategy Domain: Community/Population Health 2018 OPTIONS FOR INDIVIDUAL MEASURES:

More information

Chapter 2: Identification and Care of Patients With CKD

Chapter 2: Identification and Care of Patients With CKD Chapter 2: Identification and Care of Patients With CKD Over half of patients in the Medicare 5% sample (aged 65 and older) had at least one of three diagnosed chronic conditions chronic kidney disease

More information

Improved IPGM: Demonstrating the Value to both Patients and Hospitals

Improved IPGM: Demonstrating the Value to both Patients and Hospitals Improved IPGM: Demonstrating the Value to both Patients and Hospitals Osama Hamdy, MD, PhD, FACE Medical Director, Inpatient Diabetes Program Joslin Diabetes Center Harvard Medical School, Boston, MA Cost

More information

Functional Outcomes among the Medically Complex Population

Functional Outcomes among the Medically Complex Population Functional Outcomes among the Medically Complex Population Paulette Niewczyk, PhD, MPH Director of Research Uniform Data System for Medical Rehabilitation 2015 Uniform Data System for Medical Rehabilitation,

More information

National Stroke Association s Guide to Choosing Stroke. Rehabilitation Services

National Stroke Association s Guide to Choosing Stroke. Rehabilitation Services National Stroke Association s Guide to Choosing Stroke Rehabilitation Services Rehabilitation, often referred to as rehab, is an important part of stroke recovery. Through rehab, you: Re-learn basic skills

More information

Surgery Estimate of Usual & Customary Billed Charges

Surgery Estimate of Usual & Customary Billed Charges Surgery Estimate of Usual & Customary Billed Charges SAMPLE REPORT For more information or to order an estimate please contact: Mark Guilford m.guilford@accumedintel.com (303) 495-5191 Prepared For: Glenn

More information